Mailing List Archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [tlug] Database frontend in Linux

Thank you indeed for all the comments. It seems I have a lot to learn here, more than I had anticipated.

> You don't want a database, you want a search engine.

Thanks, Josh. You mean like Beagle? But I would want to search limited ranges of files, not everything on the computer.
> I would assume that askSam is just such a search engine: it indexes
> your files when you add them, and probably has some sort of scheduled
> index build to cover changes to the files it has indexed. When you
> query askSam, it probably queries the index and returns useful
> documents, and probably then gives you sections of those documents
> that are relevant to your query. Is that about how it works?

That's pretty much it, Josh. It does let me browse the files, and assigns them file numbers so that if I <go to> or <jump> or however that is expressed in Linnville I can go directly to a specific file. It does import/export the files, and stores them in a single bundle (per database) as an "~.ask" file. I create specific databases, one I have I just call "Health" and then open "Health.ask" when I want to search or browse those files. When I want to move a database file to another computer, I don't need to copy a bunch of folders, just the one file, and askSam will open it on any computer. It also takes care of portraying the wysiwyg view of printable documents, including graphics.

> Like Josh i think you need a searchengine. Having big binary
> chunks as objects in real databases wouldnt help you.

Pardon my greenhorn status, Christian, but I'm afraid I don't understand the difference. I do know that a relational database is way more than I need, but beyond that, I'm lost. The askSam program strikes me as being somewhat akin to a wiki, although it will handle more than pure text.
> > [ collection..]
> I use ampache for this. Its indexing musicfiles into database in
> its back. Music searching/listening is done via webfrontend.
> Amarok as standalone desktop application could also help

Thanks, Christian. I have been using Kaffeine, as it is the only program I have been able to get to work on all my Linux boxes. With the others I've been able to get success on some boxes but not others, dunno why. Totem insists on offering to search for the relevant codecs but never comes up with anything, although Kaffeine seems to already have the codecs. My main home computer is 64-bit, the others are 32-bit, dunno if that matters but I have the relevant 32-bit and 64-bit OSes installed, and I'm running SUSE 11.1 with Gnome on all of them, except the brand-new 64-bit laptop that I haven't got around to installing a real OS on yet.

> Not if we can help it. ;-) I've heard of it, but most of us avoid
> Windows and its apps as though they were the plague.[1] You really
> should describe what you're doing in more detail.

It was my intention to have done so, but I didn't realize there would be that much separation between what is going on in the-OS-that-shall-not-be-named and Linux. I bought askSam back in the early 90's, before I could possibly have been running Linux, back when it was the number one selling database program in The World According to Redmond. Back when the first Thursday of the month meant pizza night with the PC crowd in Omotesando.
> What do you mean by "tossed the results into askSam"? What results?
> How do you "toss" them? How do you access them?

Sorry to be obtuse, Stephen. I create a single file, such as "Health.ask" within askSam and import all the related documents. By "toss them" I meant that I import a batch of files at once. I generally import them in chronological order for each source for easier browsing, although I usually use the program for searches. E.g., if I want to see everything on "Bell's Palsy" I open askSam, then open the "Health.ask" file, and type in "Bell's Palsy" on the search prompt, and can generate a report or flip through the documents and copy the interesting bits, or refine the search to specific sources, etc., the types of things one generally needs to do when referring to that type of information. Of course, I can also do global searches or single document searches, or search through the results for something else, such as "Lyme Disease", as Bell's Palsy is often an early symptom of that.
> You haven't specified any requirements except "like askSam", which
> means nothing to me, and "imports text, HTML, email, graphics, and
> PDF" which isn't much more help.

I'm afraid I don't know how to explain it other than to say it imports these documents into a single database that I create. Again, pardon my greenhorn status.
> If Josh is right, and what you want is an indexing/searching
> application (he calls it a "search engine"), FreeWAIS is an
> implementation of the (now defunct AFAIK) Wide Area Indexing Service
> protocol, and I know a couple of people who use it to index email and
> other personal documents (originally it was designed for library use).
> Other well-known indexers are Namazu and Xapian. However, AFAIK all
> of these are basically libraries. They have sample front ends but
> usually adapted to providing indexes for web sites. I doubt any of
> them can parse PDF; I know Xapian and Namazu can handle HTML, though.

This is beginning to sound a lot like askSam. which can also handle ~.eml files. So what I should be looking for is an indexing app that creates libraries? The ability to handle Japanese would definitely a plus, as I do a lot of work in Japanese.

> None of these use powerful database engines as backends. Rather, they
> use application-specific databases for the index files.

Again, this sounds a lot like askSam.
> If you want something with more structure, then a relational database
> (like Oracle, Dbase II, or FoxBase) may be useful

I didn't need a relational database back when I got askSam, and AFAIK it would still be overkill for my needs. I just wanted something to store all of this information in one place, access it easily, and transport it easily.

> Two other open source databases (servers) worthy of mention are
> Firebird and SQLite, both of which have their uses as backends for
> single-user desktop-style applications.

The single-user desktop-style app describes my interest range exactly. But from what has been said above, it may be overkill.
> Unless you want to learn about DBs and indexing...
> (free, online,knows about pubmed)
> (firefox plugin)

Thanks, Ivan, I will look at these.

> The key is to find him software that will increase beer production ...
> say, a nomikai's worth every couple of months ...a
> (Ralph, good to see you finally made it over here.)

Thanks for the recommendation, Charlie. And are you going to the next nomikai? I see there is one on June 12, a Friday! Unlike the Saturdays of yore, I could actually make it on a Friday.


What can you do with the new Windows Live? Find out

Home | Main Index | Thread Index

Home Page Mailing List Linux and Japan TLUG Members Links