Hi,
On 03/08/2012 15:00, Simone Tripodi wrote:
¡Hola!
Also remember that if we ever want to deal with, say, multiplications,
monoids are only going to be in the way (we already touched this topic
before, see [1]). I'm still happy to update and simplify names, only
following a different pattern: e.g. from "DoubleWeightBaseOperations" to
"DoubleOperations". And I'd also replace "Monoid" with "Addition".
yeah thanks of the reminder - I was searching for it in the mail
archives and didn't find it :P
wouldn't "Multiplication" have exactly the same methods signature of
"Addition" aka Monoid? I wouldn't replicate stuff just to implement
markers...
Addition would have signatures like "sum" and "negate", while
Multiplication would have "multiply" and "invert".
Anyway I agree that algorithms need specific monoids, such as Dijkstra
that needs Addition - guess it wouldn't work with Subtractions :P
What about having Monoid with package visibility and then
"Addition/Multiplication... extends Monoid" ?
Then it would become a bit painless if a class had to implement both
interfaces (the current "Integer[...]Operations" is an example). I'd
just have them fully independent from each other, without a common
ancestor (Monoid).
After thinking a bit I'm also a bit perplexed about renaming "builder" to
"connect", and in general about the name of the method "connect()". You
know the meaning of "connected" in graph theory, while with our method we
could actually create a graph which is not connected (e.g. one with no
edges at all).
agreed!
So I suggest to look for a less ambiguous alternative:
"populate" (this gets my vote)? "declare"? "construct"? "assemble"?
+1 to "populate" (and related class renaming?)
cool. Class name = "GraphPopulator"? Though it sounds sooo bad to my
ears... Maybe a mother tongue can help us with the matter :-)
Ciao,
Claudio
thanks a lot for your feedbacks and enjoy vacations!
-Simo
http://people.apache.org/~simonetripodi/
http://simonetripodi.livejournal.com/
http://twitter.com/simonetripodi
http://www.99soft.org/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
--
Claudio Squarcella
PhD student at Roma Tre University
http://www.dia.uniroma3.it/~squarcel
http://twitter.com/hyperboreans
http://claudiosquarcella.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
For additional commands, e-mail: dev-h...@commons.apache.org
