Ideas, suggestions, & bug reports for foo(book):

Plain Text

Public Domain

This seems like the best place to put this.

Download (right click, save as, rename as appropriate)

Embed

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
- It should be possible to browse by popular tags.
- Users should be able to see how many people (and possibly which people) have favourited their individual scraps.
- The page for a scrap should list the name of the user who posted it, hyperlinked to a list of all of that user's scraps.
- Styles should not have to be set when creating a new scrap.  They shouldn't even be controlled by the one who posted the scrap.
Instead, individual users should be able to specify in an account preferences setting what style to use for displaying all scraps that
they view.  (Visitors not logged in are stuck with some default style).

BUGS
- Certain Unicode characters in scraps are converted to question marks by foo(book):.
 - This affects all Greek, Cyrillic, and Hebrew letters, possibly among others.  No Latin-1 characters seem to be affected, nor,
strangely, are U+2022 (•), U+20AC (€), U+201C (“), or U+201D (”), possibly among others.
 - This happens even if the syntax is set to "plain text" (which should not require any text processing and thus not potentially
mess up the text encoding).
 - The question marks are present on the normal scrap viewing page, in the scrap download page, and even when editing an
affected scrap.  These are actual question marks, U+003F.
 - This occurs when scraps are submitted or viewed using either Safari 4.0.3 or Firefox 3.5.7, both on Mac OS X 10.5.8.
Question marks are even returned when accessing an affected scrap download page via telnet (i.e., this is almost certainly not a browser
issue).
- The page for downloading a scrap converts special HTML characters (<>&) into HTML entities, yet the page is not an HTML page,
so this should not be done.
- The charset for the scrap download page is not set by its headers, and HTTP's default Latin-1 encoding is inappropriate for interpreting
Unicode/UTF-8, the most likely format that non-ASCII characters will be submitted in (and, as the edit page declares a charset of UTF-8,
all text submitted to foo(book): might have to be in UTF-8, though I'm not entirely sure on that point).
- The headers for the scrap download page list its Content-type as 'text/download', a nonexistent MIME type which the aforementioned
version of Firefox does not handle well.  If you want to force browsers to download a file rather than display it, I believe the correct
way *might* be via the header ``Content-Disposition: attachment; filename="suggestedFilename"''.