It Is A Mystery…

It’s 2008.  Why the heck doesn’t my computer know what I’m probably going to be doing when I download a file?  My computer should be able to learn that I transcode video files with English hardsubs to MP4 and put them in my iTunes share, or that PDFs that I read may tend to be related to research and should be saved…  So why doesn’t it?

I suppose it’s about time for me to announce a status report of what I’m up to lately…

First: I’ve picked up my Pixonomy project again, and while I’ve JUST put it on hold again, I’ve progressed the library with a refactoring and I just need to do some cross-platform hacking (to get it to compile nicely on OS X as a universal binary), and implement a couple of search functions to actually get it to a state where I can actually start programming client software in GTK+ or wxWidgets (I haven’t decided which) to demo the library.

Second: I’m currently taking a break from Pixonomy to work on a nifty font for OpenTTD.  Since I noted that they finally implemented TrueType support and Unicode support in 0.5.0 (which I’d tried to implement before, but never really got around to), I figured I’d try my hand at something fun.  After I get all the lower case characters done preliminarily, I’ll start adjusting the bounds and kerning by testing in game…

Third: Been watching some of the subs for this season, and I think Allison & Lillia does seem to have some promise, but we’ll see where it goes.  Zettai Karen Children, though, is not so much up my alley.  We’ll see where the other series I want to check out go (namely, Library War)…  There’s a few others that might be good, too.

I’m currently playing around with the web-based RSS aggregator Lilina to act as a LiveJournal-less Friends page, and unfortunately, it leaves a lot to be desired. It does show promise however, but it needs a lot of polishing to get there.

One reason why I wanted Lilina rather than some other source is that I felt that I wanted a personal friends page that didn’t depend on LJ, should I end up transitioning entirely off, and secondly, no WordPress plugins currently do what I want (every one I’ve seen more or less adds ‘crosstalk’ by adding posts directly to this blog, or as a simple aggregator that JUST lists article titles).

Lilina, thus far, seems to be just right: a standalone program that is lightweight enough to act as a public friends page. However, it’s clearly very much alpha, and has issues to this end. Despite this, it seems to be the best available: Feed on Feeds is more like a desktop client than a lightweight browser, Gregarius is incredibly slow, and Zort doesn’t interleave articles from different feeds.

I’ve found quite a few minor bugs in just playing around with it for thirty minutes, and would definitely not recommend it unless you’re willing to get the latest SVN code and tinker with it to make it work (for one thing, it has a fundamental flaw that prevents Slashdot, of all things, from being successfully aggregated, and most probably many others as well).

I’ll keep tinkering with it perhaps, but I might just end up writing my own app, since Lilina’s design is a bit clunky.

It’s also reminded me that I need to upgrade the default installs of PHP and MySQL on this server up to 5 (for each)…

I call it Big Project 1 because it was the first big project idea I had. For those of you in the know, Big Project 1 is the weather program (Big Project 2 is something entirely different, but still a program).

In any case, Big Project 1 has finally restarted after about 6 months of downtime. I finally perfected searching, so it actually only takes about 3-5 seconds to search for a city online, and set it as your default city. It now also has a Find City feature that automatically recenters on another city and updates the location data.

There are still a number of concerns however:

1. Searches are still not text-friendly. i.e. “Albuquerque” will return useful results. “Albuquerque, NM” not only will not, but will not even filter the “Albuquerque” results for New Mexico. This, thankfully, can be tweaked outside of the source code for the program.
2. More than two searches and certain weather data won’t be loaded. This may be a side-effect of the fact that I’m not yet using my cached weather data. Next step is to tune the format of the weather data I’m currently caching, so that I can not only send and receive more useful data besides current conditions, but also so that I can carry the burden and not have issues with FTP timeouts/numbers of sessions (which could be causing the current problem)
3. I’ve got to find sources of information for most of the useful bits (e.g. forecasts and history), especially for non-American sources, and then feed it into my custom format.
4. Some of my graphics are a bit concerning. I’ve got to not only design them, but… Well, if I detail THOSE OTHER issues here, it might be giving too much away… Suffice to say though, good 3D and imaging algorithms would be a big boon… *stares at Sarah*

But at least some distinct positive progress is being made on Big Project 1. Big Project 2 (started last month) is on hold while I refocus on Big Project 1, but rest assured that it’s standing pretty well as well. It’s already got most, if not all, of the kinks in the tagging done, but now it needs a bit of database work with SQLite, and then, lastly, the GUI design… But it’ll probably be done quicker than Big Project 1, were I focused on it.

Meanwhile, smaller projects like Ierukana’s redux are also on hold, but will likely be so for a lengthier period of time. (Nevermind that it’s serving as a test ground to learn Rails anyway).

That’s all for what I’m doing computer-wise for now… Other projects might come along that are more work-related and school-related, but we’ll just have to wait and see.

So I finally bought that iMac I’ve been waiting for all summer long. Since the 24″ came down in price, I splurged, and now it’s going to be coming with me back up to school. It’s been far too long since I had a desktop…

Is it possible to mass-produce a low-spec computer with wifi and a 3×4″ color touchscreen to retail for $50?

Unfortunately, as I was thinking about possibly turning my idea into a website, I decided to look online for similar ideas, and I found a similar website that does 80% of what I wanted to do with my program. Unfortunately, I can’t think of anything significant to differentiate my program/website from others so I may have to scuttle it, despite the time I’ve put into it.

It did come up with several useful services in the end though, but I can’t really make money off them. I’ll have more info on them as they firm up though.

Mostly for future reference as I don’t have the time right now…

An RDF Inspector and SPARQL interface for Firefox plugin?

The more observant of you may have noticed some signifier of the FOAF linkage on my front page (returned after a hiatus). That’s right, you know who I am now!

That being said, everything up to now (and my quietness) has a fairly long backstory.

It started back in November of last year, when an attempt to renew pipian.com went horribly wrong (Thanks RegisterFly!) $130 and nearly two months later, I finally got it back and, by that time, had started on a new project.

This project is still somewhat secret as it’s incomplete at this time, but it’s progressing well (and some of you already know of it) and will hopefully will be marketable by the summer (crossing my fingers of course).

In the process of working on this and getting back into the groove at school (and in all the hecticness of setting up and being Chief of Operations at Genericon XX) I managed to finally (and briefly) meet Jim Hendler, who is currently in the process of moving to take a constellation chair at RPI this semester. This got me thinking.

One of the items that my ’secret’ project needs is something of a stable geospatial framework of coordinates, cities, and hierarchies, to enable intuitive and ’smart’ discovery of as many cities as possible, and as needed. Thus, I came up with the concept of the Semantic Web Locationary, utilizing several well-known semantic ontologies and several free/libre data sources to accurately describe many geopolitical constructs (mainly hierarchies of city->province->country->continent->Earth styles).

I figured this would both help to get my feet wet in a seriously usable semantic web context (The Geonames.org ontology is not terribly human-friendly, even if machine-friendly and more detailed than the Locationary) as well as offering a framework for my program to rest on as well as offer a static reference for other semantic applications (foaf:basedNear anyone?).

At any rate, it’s somewhat stable now, in so far as all the MAIN parts are implemented (if a bit in semantic flux at the moment), so I can move closer back to the project by overlaying one more data store over the existing Locationary (Arash Partow’s Global Airport Database, given the fact that DAML no longer has their airport script online, DAFIF is no longer available, and my project needs airport locations) and probably making that publically available as well…

More information about what I’m REALLY planning with this semantic data when the project gets closer to completion.

Finally, a week without much homework… After a month of almost working solid. Well-earned I think…

I might be able to get something useful done and earn some money for once too.

Also something cool I just learned: The venerable xscreensaver has a native Mac OS X port. Now I can have GLMatrix again! (Poor Windows doesn’t get it, and XGL doesn’t like displaying OpenGL in its windows)