🚀 go-pugleaf

In article <XJAM.92Dec14233911@cork.CS.Berkeley.EDU>
xjam@cork.CS.Berkeley.EDU (The Crossjammer) writes:
[...]
>The Dylan *name* is trademarked just as UNIX is trademarked by AT&T. Apple
>will have to actively protect to maintain it as such so you can expect some
>litigation down the road. The validation suite is supposedly to ensure that
>anyone using the name Dylan meets whatever specifications Apple sets out
>for the language. You'll have to meet it to at least have a shot at using
>the Dylan trademark.

The commercial marketplace has an awful lot of brand name affinity.  Many
people will shy away from any product that has a taint of 'incompatibility'
(think about the argument "Gee, if the really are compatible then why did
they have to call it something different").  This is/has been eroding to
some degree due to the enourmous market power of the 80x86/DOS world
bringing an amazing amount of competition.  Technical excellence does not a
sucess make.

>Having said all that, you miss the point of Thomas. Anybody can get the
>Dylan manual and implement a language that meets that specification. I
>imagine that the validation suite will be publically available so you can
>even see if your implementation is up to "snuff". But unless you go begging
>to Apple you can't call it Dylan.

>In my book that means the word "Dylan" isn't free, but the language is. At
>least until Congress gets through mucking up the copyright and patent laws.

Yes, but either Apple stays in the driving seat (driving the language where
they want it to go) or "Dylan-based language" explosion happens because
people don't agree with Apple's views and then you're essentially back to
resticted markets (hmm..., a reasonable reason for creating a canonical GNU
version :-).

[...]
>The current Thomas implementation isn't exactly a speed burner, but people
>can work on it, or write their own implementations, or whatever. But as far
>as I can tell, Apple has no means by which to "lock" people into their
>hardware using a language. People lock themselves in when they *must* have
>the language, but don't want to go to the trouble of implementing it
>themselves.

My (limited) understanding of the whole PDA effort at Apple is that they
have a proprietary hardware platform with a proprietary OS that just
happens to be written (to some uspecified degree) in Dylan.  They release
the Dylan language to anyone who asks (hell, they nicely mail a perfect
bound volume to you :-) so that people become aware of/familiar with the
language well before the actual platform is ready for development.  When
the platform (and more importantly the SDK for it) become available they
already have at least some market awareness.  The short of this whole line
of thought is that by dangling Dylan out to the world they get more people
to develop for their upcoming (proprietary) platform and thus generate lots
more profits than they could with tools that would require them to compete
with others on an even field.

>Technically speaking, there are some interesting ideas in Dylan and a FSF
>implementation would be another datapoint. But I doubt if there will be
>many takers.

As a research, niche type language, I think that Dylan (or whatever :-),
is fine.  I'm just saying that Apple has a long history of doing things to
sucker people into believing what angels they are and then, when it's too
late the people figure out that they have been had.  I'd just hope that
whether people agree with me or you or not, that they at least think
through some of the issues and decide themselves.

Thanks,
	John

P.S.	I'm not sure if this conversation has anything to do with this
	newsgroup.