From my new subscription to the Twine (data?) service, I subscribed to the RDFa group to hear, and read about new (or interesting) technologies around the web. It's been quite some time since I had heard about Freebase Parallax (last read about in the Web4Lib digest), and when reading over the article in my Twine digest email, I took a second look. The video demo really showed its power. But, to understand the Semantic Web (RDFa), I'd suggest anyone interested read over that article first. It discusses ideologies on how one should be able to find information and data over the web.

A Good Data Browser Allows You to Navigate the Knowledge Space by Car
...and...
Twine, if you're interested - currently in private Beta.

The title of this post is a bit misleading; I was trying to stuff it for SEO keywords without it being too long.

I ran into an issue with regard to the CSS property of "overflow" that I had set to "auto". When I set some test text of about 100 or so continuous "Hello " words, the scroll bars worked just fine. I then tweaked my node's height, and text and continued, but... The scroll bars weren't showing up, and my text was outside the bounds of the box (and therefore clipped)! Oh no! Tragedy of tragedies!

Through a little experimentation (in the end, this took me about 30 minutes to track down, and then test in all the browsers), I was able to determine what the problem was.
Continue reading "Browsers and minimum CSS overflow on Windows XP"

An interesting question came up on the Web4Lib daily-digest today. I replied directly to the author (I think, never understand how listservs work) but I thought I'd share this anyhow as it's not a hugely sought-after solution. Note: there are probably other, better ways to do this. Comments and other solutions are welcome. I am under the assumption you are using Windows XP (if anyone wishes to make alterations for Vista or another OS, you're more than welcome in the comments - if Linux and not using the shell, please specify which distro).

The question:

Has anyone tried to network an internet connection that involves a Verizon PC card in a laptop. Ideally I would like to be able to have 2 or 3 laptops sharing that one connection, but I have no idea where to start!

Hopkins County - Madisonville Public Library

My response:

You would need 2 other components:
- 1 crossover cable (it's a CAT5 cable with slightly different wiring)
- 1 "splitter", so a hub, switch, or router.

If you use a wireless router, you'll only need the one cable. If you use a hub, switch, or regular router you'll need a standard CAT5 cable for the other computers to connect to.

1. Connect the crossover cable to the LAN port on the back of the laptop.
2. Connect the other end of the crossover cable either in to a single computer, or the "splitter"'s "INTERNET" port.
3. In Control Panel, go to "Network Connections".
4. CTRL+CLICK on the two network connections you want to be sharing (the Verizon, and the Local Area Connection).
5. Right click on one of the two, choose "Bridge Connections". (I can't verify what to do to finish the bridge as I only have one connection on my PC at the moment. A quick internet search could probably tell you.)
6. If the "splitter" you were using was powered on, you may have to wait a moment for it to get an IP address, otherwise you can always power cycle it to renew its IP. The same goes for the PCs if you don't know how to force it to get a new IP.

I *think* that should do it. If I missed something from memory of when I did internet sharing, you can just do an internet search for examples or tutorials on "network bridging winxp". Switch out "winxp" for whatever OS you might be using if otherwise.

I hope this can help someone looking for a solution - or at the very least, get them on the right track. I am assuming here that the cell provider is not using the LAN port (either a USB or PCMCIA port).

Update: I believe I've enforced some stricter SPAM filtering on my blog (i.e.: any at all). At the very least, I guess I'm now more popular? Sorry for any inconvenience. Comments are enabled once again.

There was recently a nice posting from PHPDeveloper.org linking to an article by Florian Eibeck, where on his blog he discusses some solutions to these extremely fun situations.

http://blog.thinkphp.de/archives/342-Multilingual-Websites-with-PHP.html

Today is the release of Firefox version 3.0. The world over plans to celebrate the release by pledging to download the browser in an attempt to create a Guinness Book of World Records entry for most downloads of a single product on its release day. Although there are no other holders for this entry, the Mozilla Corporation intend to beat Firefox 2's record set a few years back. Join in if you are a Firefox user!

In other news, Opera v9.5 is out, and is sporting some fancy CSS3 additions!

Disclaimer: These are primarily notes for myself as I get accustomed to CakePHP's (v1.2) authentication and ACL. I've been building a website and wanted to finally make use of CakePHP before I start using it for the library's CMS. A CMS is a large undertaking and I don't want to go into it "cold-turkey", so I came up with some side-project to help me understand the framework a bit better. All in all, it truly is an extremely rapid development tool.

I decided not to go with ExpressionEngine or Drupal simply because I'd still have to take time to learn those systems and make modules or extensions within those languages - and there's no guarantee I'd be able to do what I'd need to do with them. If I build it myself, it would take just about the same amount of time with Cake. Anyway, to continue...
Continue reading "CakePHP Auth Component"

So here I was trying to do some work on my domains this weekend and into today, thinking that my host had problems with the server on the weekend... "Oh, it's up again today," I thought...so I proceeded to try to login again today. Well, apparently I forgot my password somehow and so I locked myself out from my own websites from two separate IP ranges. Awesome! At least my own place isn't blacklisted (visited my parents this weekend and was at work earlier today during my lunch break).

Ha! Oops... I still don't know what my password is for cPanel, but the regular account login still works.

So, in attempting to figure out this rather obnoxiously generically defined protocol, I've found that our vendor's implementation of the protocol is not what I would have expected from first attempts. Properly formatted XML with line breaks to denote a new tag in the structure is apparently not desired as the query fails. If I remove all newlines (and I did remove all leading whitespace, though I'd imagine it was possibly unnecessary?) the query was successful. ...not to mention that a test was given where we would access it using Telnet under Microsoft's DOS. Well... MS-DOS prompt's telnet seems to be quite finicky and doesn't work as expected. Using PuTTy seemed to work OK though. Go figure. I'm glad I have that for SSH to our host otherwise I wouldn't have thought of it.

Without an RFC to describe exactly how communication is to be sent, received, and expected...bug testing has become a chore and a bore...not to mention that in trying to get a PHP socket client working just does not seem to be working. I can read information sent from my test server, but I cannot send data to the test server; though it works with PuTTy. If only it were a webservice with SOAP or something similar instead...which is probably what I'd extend this to do anyway, eventually...for use by the consortia.

Oh yes, NCIP v2 should be coming along within the year as well since I believe it's now been passed. Yay. Give me more technical specifications or an actual library example to work from! ARGH!

I've updated the plugin version to the most recent core files (from somewhere of like August 2007...oops). If you don't wish to download the files using Spartacus, you can manually install the files by downloading them here

Those of you looking towards "widgets" and cross-platform application development, but are primarily web developers with some background (or none) of programming with a desktop application, perhaps you should take a look at Adobe AIR (formerly known as Apollo). It allows you to create rich, cross-platform applications using just HTML, CSS, and JavaScript! ...and, technically, I suppose you could leave out the CSS and JS part of that and it would still run...but what's the fun in that? It can be integrated into a Flash application, or a Flex application - but it does not have to be.

If you're seemingly interested in the technology, I went through and found some interesting links on the subject (read: tutorials) that will hopefully help to get you (me) started. Okay...so I did it because I'm interested and this blog can serve as an access point for me rather than using my Bookmarks as a temporary storage medium. Whatever. I was looking for simple tutorials, so I ignored anything with an "adobe.com" or "ibm.com" (developer works) domain, so if you'd like more when you dive head first, you might want to look to those resources instead of the ones I've provided below.

1. Pete Freitag - newest one (by published date) that I could find
2. NOT A TUTORIAL - Get the AIR SDK here
3. Jonothan Snook on 2007's 24 Ways: Christmas is in the AIR - a simple to-do list tutorial
4. AOL Developer Network - Part 1 of a Series - Simple "Hello World" type page with a self-signed certificate
5. Jonathan Snook (again) releases Snoto Foto source code to help you learn! - No tutorial, but I'd imagine it's commented well

Enjoy! I hope I will!

Update: Make sure you've updated to the latest Sun Java JRE package or you might get some really, really odd and weird undocumented errors when trying to compile!

Well it's about time, the site is now live! As I've said in a previous post about this, it's not complete - all the content (sans images...which was almost all clipart) is directly from the old version of the site, just rearranged.

Put in use to this design are some rather obnoxiously tricky CSS tweaks along with JavaScript, Flash, and RSS feeds all pulled together to create a single site.

JavaScript:
sIFR
Dustin Diaz's SweetTitles (tooltips)
Homebrew DL list FAQ
NiftyCube
AmberJack Tour Script (temporarily)
...and some other various stuff...

CSS:
Tripoli CSS "Framework"
Stuart Landridge's Image Replacement Technique

Flash:
Monoslideshow ($20 purchase, site license)
sIFR

...and lots of HTML! Ha! I also used the WeatherBug API, and MagpieRSS (I used my own SimpleXML RSS reader but I didn't write a caching method, Magpie seemed faster and it already had caching, so...), as well as a "Beta_RSS" feed of our Event Calendar (which is incorrectly serving non-ISO-8859-1 as ISO-8859-1, causing problems I can't figure out how to fix. Regardless, it looks pretty!

Now, to wait for the dust to clear and the smoke to settle (all the little things that bug me that I will fix, but aren't important) so I can move on to updating the content and more visual hierarchy of things. After that I should start talking about programming some more here.

I spent more than half the day today (24 hour day) thinking and trying to find a way to get a mod_rewrite call to work properly. I felt so dumb, retarded, and plain uneducated in my attempts. I left work a little bit early so I could concentrate (no distractions, no phone calls, no "how do I send an attachment", no nothing). Even after that, I kept running into a stumbling roadblock. My friend was trying his hardest to help me out (no idea why other than possibly a sheer curiosity - it's now almost 1:00am).

What I wanted to do:
http://example.com/contacts/form/recipients/bKozlowski/
...transformed to...
http://example.com/contacts/form/index.php?contact=bkozlowski

Sounds easy, right? Yeah, I thought so too when I first started. I tried every iteration of something that looked so darned simple and could only be done just so many ways... The eventual problem?

I needed to use: RewriteBase /
I have no idea what this directive does. I have no idea why my pattern now matches and works. The RewriteLog really wasn't helpful at all except to tell me that it wasn't working...something I could have figured out without a log, obviously. Anyway, if anyone really cares, the following is the actual end result:

RewriteEngine On RewriteBase / RewriteRule ^/?(contact)/([A-Z_@\.]+)/$ contact/form/index\.php?$1=$2 [NC,PT,L]

I both love and loathe mod_rewrite.

(Special thanks to the people over at SitePoint Forums for their immediate response!)

For the new website I've been developing and finishing up some "final touches" on the design to make the experience just a little bit better (before I can delve into the out-dated content after launch), one of the last things that was needed was to speed up the efficiency and download of assets - this included CSS and JavaScript files for the most part. I made the sIFR flash files as small as possible and my PNG images were compressed (and fixed for IE color rendering) using PNGGauntlet (which uses PNGOut). Until I can figure out ETags, a proper time for an Expires header, and GZip to work on CSS/JS files on our hosted server, this should suffice. It loads amazingly quick on my iPod Touch.

Continue reading "Compare your Minimized JavaScript"

Clarification for those of you that read my blog for non-library-related information: this is library related.

I just got an email from one of the sales associates (I somehow got added to the list) on a new product that can be incorporated into the PAC: NoveList Select.

Polaris is pleased to offer NoveList Select - a quick and easy way for patrons to find books similar to those that interest them. NoveList Select pulls from a database of over 4 million titles and retrieves only books that are included in your collection. What's more, a "Find more like this" link appears right in the PAC and sorts results by popularity, making it easier for readers to find additional books they will enjoy.

I can't recall if other PAC software does this either by default or with a company-supported plugin (I know there are third party plugins), but I'm quite happy to hear about this. I'd imagine it works under a library consortia running Polaris just fine, so now I guess I just have to hope our consortia decides to take a serious look at this.

While using Mon.itor.us to monitor the uptime of my websites' uptime, their IM messaging has not worked for me for quite some time now. They offer SMS messages for a fee as an alternative, but that's no fun considering it costs me $0.15 per SMS as it is. I just came across Pingie, a FREE service that will send an SMS to your phone when it encounters new posts from any feed. Lo and behold, Mon.itor.us provides an alerts feed. Yay.

Pingie.com