well, i’ve finished entering the suggestions from the forum into the project database, and have nearly finished those from email. please note this report is very rough, there might be duplicate entries, and the assigned versions will likely be changing. also, there are further details for each item not displayed in this report, including the detailed description. i wouldn’t say the list is exactly readable yet, but i promised that i would post updates as i went along: NeoMem Feature List by Version (63kb pdf). this will eventually become an interactive list.
(note: this also does not include many ideas that are still stuck in unformatted text, on paper, and on index cards. but it’s all getting into the database, eventually. i’m trying to balance getting this stuff entered with working on the architecture of the program.)
the amount of work required to enter this feature database has led to some more feature ideas:
- refactoring text into an object - ie if you have bunch of text, easily chop it up into an object, with properties, description, etc.
- a code view on the underlying code of an object (which would be plain text with wiki/markdown/email type syntax). for refactoring, it would be simplest to just mangle the plain text and let the program parse it out into properties etc.
- importing email (thunderbird stores your email in plain text files) - would be useful to be able to refer to the original email at times, and/or just refactor it into an object.
- importing phpbb forum entries (stored in a mysql database) not sure i’ll be sticking with this program though.
the refactoring capabilities would come in handy in converting a webpage into an object, which i’ve been doing a lot of lately in researching related projects and tools. for instance, you could copy and paste a web page into the editor, then chop it down to something like the following, which neomem would parse out into fields and store in the database:
name: Commons Virtual File System
aka: Commons VFS
url: http://jakarta.apache.org/commons/vfs/
desc: Commons VFS provides a single API for accessing various
different file systems. It presents a uniform view of the files from
various different sources, such as the files on local disk, on an
HTTP server, or inside a Zip archive.
Some of the features of Commons VFS are:
* A single consistent API for accessing files of different types.
* Support for numerous file system types.
* Caching of file information.
* Event delivery.
project: neomem
rating: 4
added: 2005-10-25
might need to keep this text for reference as well, especially if there were bits it couldn’t parse or understand - i’d like neomem to be a sort of fuzzy database, extracting as much information as possible from possibly ambiguous information, asking for clarification where needed, and keeping the original text around for reference.
[and of course, it would be simpler if websites would provide information like this. ie provide a human view of something (html) and a machine view (xml or rdf or some variant like this wiki/email/markup code).
update: there is such a thing! Resource of a Resource - this looks really cool - spread the word. much cleaner than embedding it in html with meta tags.]
Tags: No Tags
5 comments
Comments feed for this article
October 27th, 2005 at 7:26 pm
Kevin
Cool ideas. Any further discussion on the database backend? Are you going to use Access? How about SQLite, or some other variant. mouser, at www.donationcoder.com used a (shareware) database component for his clipboard program (CHS), it looks pretty powerful.
Kevin
October 28th, 2005 at 12:04 am
ivan
imho, MySQL as backend is a very nice choise due to its freeware status and popularity.
October 28th, 2005 at 8:32 am
bburns
i’d like to get away from proprietary products like access, so probably mysql or sqlite. but i’m thinking the backend will become a virtual file system, with plugins for different data sources. so it should be able to support any data stores, including various databases, the file system, or zip files, etc.
it wouldn’t be much work actually, just need a layer of abstraction between the data and the backend. and it would be a way of providing metadata for the file system, something which is sorely lacking. more on that later….
thanks for the pointer to donationcoder.com - interesting stuff there.
~brian
October 28th, 2005 at 9:21 am
Kevin
Sounds like a good idea about the backend. Especially being able to support different data types. Experience from the Keynote forums shows just how many different ways people use programs like this. Text editing to random data storage.
Yeah, donationcoder is a tremendous site. Why dont you go announce yourself on the forums there? Mouser is extremely helpful and already is aware of neomem. In fact, there are talks about trying to turn CHS into something like neomem. It would be awesome to combine ideas into one program. My forum name is kfitting over there.
Kevin
October 28th, 2005 at 10:39 am
bburns
yes i don’t know why it took me so long to hit on that idea for the backend. it’s all about plugins…
thanks, i’ll do that in a bit -
brian