Potential Network SIG

[Johan Tibell]

On Mon, Aug 24, 2009 at 10:43 AM, Yitzchak Gale<gale at sefer.org> wrote:
> Johan Tibell wrote:
>> We shouldn't be afraid to use composition to
>> prevent a combinatorial explosion i.e. don't have a sendAsUtf8,
>> sendAsUtf16 but instead do
>> send $ Text.encode Utf8 "myString"
>
> OK, makes sense. But then, that should appear very
> explicitly in the documentation. It will be the most common
> usage, and people using it will not be focused on Network,
> Bytestring, and Text, but on other things. They should not
> be forced to thrash around and do a research project just
> to do something simple.

I agree. The network package could use a lot more documentation in
general. For example, I put a echo client/server pair example in my
network-bytestring package as the network library lacked even
something that basic.

> Thomas DuBuisson wrote:
>>> 6) Integrate with the rest of hackage.
>>> This means instance of PrettyClass, Parsec, and Binary.
>
>> I must disagree here. We don't want hard coupling to lots of other
>> libraries.
>
> It won't be a problem if the convenience glue is in separate
> packages that pull in all of the dependencies. But then, these
> other packages need to be mentioned in the documentation
> for the main package.

That is an option. Perhaps something we could look into once we dealt
with the basic design of the network library itself?

-- Johan