*gulp*

Feb. 2nd, 2004 06:55 pm
katster: (academics)
[personal profile] katster
Still haven't come up with a final project idea. And the clock is ticking. (The one's I've got are neat, but I don't know if they're up to my skills.)

Whee.

http://www.sims.berkeley.edu/academics/masters/final_project/ -- the last little while of final projects can be found there.

[Edit 8:52 PM: Lemme add some clarification to this: It has to meet these criteria. It's gotta be something that I could get done on my own in a little less than three months without forgoing eating, sleeping, or my other classes. (And I'm not much of a programmer.) And it has to deal with some aspect of information management. So, given these criteria, is there some sort of web application that you would really really want?]

Date: 2004-02-02 07:09 pm (UTC)
From: [identity profile] tangaroa.livejournal.com
So, what are the ideas?</interested>

Re:

Date: 2004-02-02 07:12 pm (UTC)
From: [identity profile] katster.livejournal.com
The ones folks have suggested here: http://www.livejournal.com/users/katster/382228.html

But I'm not sure if my programming fu is up to it.

-kat

Then try the programming bar ;)

Date: 2004-02-02 07:46 pm (UTC)
From: [identity profile] tangaroa.livejournal.com
Hmmm... From what I've heard, I believe [livejournal.com profile] hitchhiker's suggestion is called Microsoft Sharepoint Server. That might be out of your range. You'd have to know what to build each individual module on, and how to connect them together, and so on.

[livejournal.com profile] siliconshaman's idea interests me because I'm a GIS nerd. You would need to get the GIS data in the first place, and then have special fields for the data table that are used by the search program, for instance, every object has a field called Name. Then google for "fullname" + "region", each individual word in fullname, and so on, assign some kind of ranking mechanism, and present. I'm not very familiar with GIS programming options. UMN Mapserver can be programmed with, but it's set up to reload from file every operation, making it slow. Still, it will work. [livejournal.com profile] slipjig's added suggestions bring to mind the idea of making [parts of] the database table user-editable.

[livejournal.com profile] rbos's idea is interesting, mainly for the fact that it's just over my head and I'd have no idea where to start with it.

[livejournal.com profile] tamnonlinear's idea is a bit vague... If I'm reading this right, aren't the job announcements themselves telling you what you need to get into the position? Perhaps a job announcement database? more strict parameters are needed, since I don't know where to start implementing something like that. If it's not going to be database backed, someone could just throw up a wiki, and that's not really much of a project.

I initially misread that as suggesting a database of user skills, and thought, Sourceforge already kind of lists programmer experience. Something different I've imagined before is a way of confirming a user's experience, like making sure someone who calls themselves a C expert can write a pointer to a function. The difficulty there comes in figuring out how to automate questions in a way that people can't cheat on it.

In sum, I like [livejournal.com profile] siliconshaman's and [livejournal.com profile] rbos's ideas the best.

Re: Then try the programming bar ;)

Date: 2004-02-03 12:53 am (UTC)
ext_74: Baron Samadai in cat form (Default)
From: [identity profile] siliconshaman.livejournal.com
Hey thanks !! For a rip-off of one of Willam Gibbson's ideas it's not bad. [I only pilfer from the best!]

Hey Katster, you could maybe make it easier for yourself if you use Google's API's for the core of the search engine. [check out the google art today too, seriously cool!]

Re: Then try the programming bar ;)

Date: 2004-02-03 08:38 am (UTC)
From: [identity profile] katster.livejournal.com
It's a thought. I'm actually sorta leaning towards a neighborhood guide, but I'm trying to play around and get my head behind an idea. Nothing's really sparked the "oh my god this would be cool!" button, but...

As for the google art, it's pretty cool. :)

-kat

Party Pooping

Date: 2004-02-03 12:39 pm (UTC)
From: [identity profile] shmuelisms.livejournal.com
I'd rather not be a gate-crasher, but I must point out that the data needed is more complex than one would initially think. You would need to know, for instance, the exact shape of each building, in order to calculate line-of-sight stuff, just to be able to construct a 3D model, in order to identify the buildings being viewed.

I would find it extremely doubtful this info could be easily had in the Land of Home Security Inc. I say this, so dear Katster doesn't get too sunk in this project too easily.

Best of luck, to you, yo Kat!

Note

My main blog is kept at retstak.org. I mirror posts to this Dreamwidth account, so feel free to read and comment either here or there.

November 2020

S M T W T F S
1234 567
891011121314
15161718192021
22232425262728
2930     

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Jan. 16th, 2026 05:44 pm
Powered by Dreamwidth Studios