From mboxrd@z Thu Jan 1 00:00:00 1970 MIME-Version: 1.0 In-Reply-To: <8cee7204b91c9558c86347e572df6ad5@coraid.com> References: <5db2df69e768a7259b8a1e698abddd3f@coraid.com> <32125c2033c0f63e5bb24548c7bc934f@hamnavoe.com> <3e1162e60909021127m7c22b518hf2255495dce5ee30@mail.gmail.com> <8cee7204b91c9558c86347e572df6ad5@coraid.com> Date: Wed, 2 Sep 2009 11:46:50 -0700 Message-ID: <3e1162e60909021146ieae36cex89d66e2d527354b1@mail.gmail.com> From: David Leimbach To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net> Content-Type: multipart/alternative; boundary=000e0cd4cc0a98f2a504729cb0d9 Subject: Re: [9fans] nice quote Topicbox-Message-UUID: 5fa001ee-ead5-11e9-9d60-3106f5b1d025 --000e0cd4cc0a98f2a504729cb0d9 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On Wed, Sep 2, 2009 at 11:35 AM, erik quanstrom wrote= : > > > > on p. 43/44 i believe it is claimed that one > > > > cannot do CSP without pure functional > > > > programming. > > > > > > (p =E2=87=92 q) =E2=87=8F (=C2=ACp =E2=87=92 =C2=ACq) > > > > > > > > That's interesting because pure functional programming doesn't exist at > all > > in the strictest sense on a computer. One MUST be able to cause side > > effects during computation or your CPU will just get hot... if even tha= t. > > i read the slides as contrasts, not as > logical conjunctions. > > i still don't understand the claim that message passing > requires "thousands of message protocols" > and can't do syncronization. > > I also don't get that. What was meant by his usage of "protocol". Erlang uses only a handful of patterns that work really well for interaction in each subsystem. If they think of messaging and protocols in a smalltalky way, then each class has a protocol of messages (methods) that must be implemented, but I don't get why that's bad. It's called an API. I mean HTTP has a small protocol, but if you count all the things you can d= o with REST, then it looks like a lot more. Dave > - erik > > --000e0cd4cc0a98f2a504729cb0d9 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable

On Wed, Sep 2, 2009 at 11:35 AM, erik qu= anstrom <quanst= ro@coraid.com> wrote:
> > > on p. 43/44 i believe it is claimed that one
> > > cannot do CSP without pure functional
> > > programming.
> >
> > (p =E2=87=92 q) =E2=87=8F (=C2=ACp =E2=87=92 =C2=ACq)
> >
> >
> That's interesting because pure functional programming doesn't= exist at all
> in the strictest sense on a computer. =C2=A0One MUST be able to cause = side
> effects during computation or your CPU will just get hot... if even th= at.

i read the slides as contrasts, not as
logical conjunctions.

i still don't understand the claim that message passing
requires "thousands of message protocols"
and can't do syncronization.


I also = don't get that. What was meant by his usage of "protocol". = =C2=A0Erlang uses only a handful of patterns that work really well for inte= raction in each subsystem. =C2=A0If they think of messaging and protocols i= n a smalltalky way, then each class has a protocol of messages (methods) th= at must be implemented, but I don't get why that's bad. =C2=A0It= 9;s called an API.

I mean HTTP has a small protocol, but if you count all = the things you can do with REST, then it looks like a lot more.
<= br>
Dave
=C2=A0
- erik


--000e0cd4cc0a98f2a504729cb0d9--