>
> But if it exists it should be fairly easy to revise and extend it -
>
> maybe someone will will want to.
>
>
I agree. The fact that they have a C-API and a Tcl binding (I have used it briefly and it works) for testing purposes proves that it can be revised.
>
> As for Redis, it's a key-value store, and it might be a good one. At
>
> least it's not claiming to be a database.
>
>
They do call it a DB and it supports logging and snapshot-ing (like prevaylor) - to go by their documentation. But I am not an expert to say that it covers all use-cases to be a robust DB though.
>
> I do not understand why Mr Falcone (or whoever he is) has such a thing
>
> about Prevayler, which he mentions in one-line posts at almost regular
>
> intervals. Are they paying him?
>
>
No comments. But, however, I dont view Mr Falcone's mails with a not-so-positive view. I view him as a well-wisher of Tcl who wants Tcl to be not just a language...but a platform providing/scaling to various applications/use-cases. To be frank, his one lines speak a lot to me.
>
> Both he and Prevayler themselves (and others) seem to be stuck on two
>
> misconceptions:
>
>
>
> * "we can do this and it's fast and cool so we should use it for
>
> everything!"
>
>
>
> * "old = bad, new = good"
>
>
>
> The inevitable backlash against this may or may not involve a revival of
>
> the Relational concept, but it should be fun to watch, and will
>
> inevitably be a rediscovery of the issues which led to the creation of
>
> relational databases in the first place.
>
>
>
> Eric
>
> --
>
> ms fnd in a lbry
I share your thoughts based on my understanding that Prevayler has not been adapted by many (or has not become popular or a standard) yet. But, it seems to be under active development....as they have new releases made up until 2013.
But, if its good... let it happen to Tcl also... is what I expect as a Tcl programmer.
Regards,
nagu.