[circle] Many notes on my Circle experience

[Asheesh Laroia]

Quick warning: The following paragraphs are largely unrelated.  Treat them
as separate messages, by and large.

In my environment, we decided there was no legitimate reason to get a
user's IP address, and that all the users were in the same time zone, so
we're going to quietly remove those from the "Display User Info" dialog.

During hashing files on Windows machines, the GUI becomes entirely
unusable.  Once file hashing is done, the GUI is usable (and the glory of
Circle is available!).  I'll talk a bit more about the hackery I
resorted to to get a usable win32 client in another email.  I'd like to
have less hackery and more responsiveness, anyway.

There was some talk of the profiler; if someone could tell me how to run a
the win32 client under the profiler, I would be very happy to do this.

Some users want the ability to search by directory, since many data
formats lack good metadata support (and there are some poorly-tagged MP3
files in the wild).  I see a moderately easy way to do this as one of the
following:

        * Add directory keywords to the keyword list of each file
        * Add directory keywords in an established metadata field all
files share (call it "Directory Name: /subdirectory/of/shared" or
something) and make that searchable the same way MP3 metdata is

Granted, I have no real experience with this program, so what I think is
easy and what you think is easy may vary.

Sometimes in Windows, when running Circle as a .pyw program (one that
doesn't have the Python terminal window in the background), a
command-prompt box comes up to let Pango complain about a non-UTF
character.  If this is not a known bug, I can find more information about
it, too.

Some users want a search query history, like many Gnutella clients have.
If you're not familiar with the concept, the idea is a concise table or
list of what things have been searched recently so we can all laugh at
ourselves.

You use "their" a lot when a singular pronoun is better.  Now, for our
local distribution, I'm going to change almost every "their" to "user's"
or "peer's".  I'm slightly wary of starting a prescriptive-vs-descriptive
grammar flamewar here, but I will offer that if this is desired by you
guys, I will contribute back this [trivial] patch.

What would be the easiest way for clients to determine the size in files
of the network, if we can assume that never will more than 400 people be
connected?

And can file hashing be serialized or saved SOMEHOW between starts of the
program?  Perhaps based on a combination of file size, modified-time, and
name.  We are unlikely to have people TRY to defeat this in our network,
and we could have random file rehashing perhaps.

(That turned out to be longer than I expected.  Ah, well.)

-- Asheesh.

-- 
In a consumer society there are inevitably two kinds of slaves:
the prisoners of addiction and the prisoners of envy.