Category: Technical

MS. DEWEY
http://www.MsDewey.com
Meet the world’s more sarcastic bitch of a search engine. Technologically, she is a step up from Jeeves, but she is a bit more witty and insulting. This is a search engine for the thick skinned, or anyone willing to learn what searches are safe with his drama queen.
Some of my favorite searches to get a personality profile of Ms. Dewey (Ms. Thang) are:
1. Gangster (bitch)
2. Gangsta (insightful)
3. Fetish (creepy)
4. Beer (funny)
5. Viagra (super villan)
6. Spam (tortureous)
7. Masturbation (bitch)
8. Ms. Dewey (disfunctional)
9. books (random)
10. depression (evil)
11. suicide (humor/disturbed)
12. strip (creepy)
13. country (really bad impressions)
14. big bang (overly sexual)
15. clone (MPD)
As a gay male, I find many of this slut’s perversions to be a bit distasteful.

Is anyone shocked that this post is not about ceramics?
I am straining my brain working on GITI v2 AND GITI SiteEngine’s permissions systems. I feel like this is a system that has been done to death in the form of NTFS, .htaccess, Passport, LDAP, etc, etc; but yet, it is essential that I write this from the current point of GITI authentication in order to allow SiteEngine and GITI to both work properly when they are introduced to more users. This is less important for GITI, as I am about its only user, but with the places im putting SiteEngine and the ways it will be used, it is becoming more important that it be able to authenticate users and that their realms/perms be respected. Some time ago I decided that GITI and SiteEngine would operate on separate, but equal authentication realms. GITI and SiteEngine users can authenticate to SiteEngine (the subordinate product), while only GITI users can authenticate to GITI herself (controller of all that is technological or intellectual). These permissions tables and their rules have been understood for some time, but at first I started assigning roles (much like those used for Microsoft SharePoint Services), giving no thought to user-level microcosm control. In a way I am no closer to that now than I was before I started this endevour, but then again, maybe I am? Currently I am working on the concept of feeds (since thats what the new site will need to understand for access control). The concept for control is that no user can add themselves to a non-public feed, and all users will be assigned to the private feeds they are allowed to access, however, they will be able to turn them on and off at will.
Maybe I should sleep before I start having nightmares about the code i’ve written and the database tables i’ve created.

I am currently working on modifying my personal website (http://curtis.kularski.net) and have decided that my current domain (curtis.kularski.net) is too linked to its upper level domain (kularski.net). I want more freedom from this. Do any of my readers have any suggestions?
If someone can come up with a domain that I like enough to use, the winning suggestion will win their choice of handmade ceramic bowl (or the option for a custom bowl). The rules are simple, I need a domain that reflects who I am. It shouldn’t be too long, but it must be available in the global DNS (you can use http://www.whois.sc to confirm).

The two most downloaded images from my blog are the following:

I don’t get it. I understand the pic of the guys kissing (sort of), but whats with the freaks who are posting the pic of the baguette to their profiles? Today when I started my log excavation project (curiosity&&boredom driven), I started to get pissed at the number of people who link to these images instead of downloading them and using them on their own space, but then I remembered that not everyone has the resources I do for such things, as well as got a chuckle out of the number of people who are that interested in images that I’ve used in past blog posts. None of the other images I have ever used on the blog come close to the numbers of hits that these pics get on a daily basis from MySpace and various forums.

I was just having a merry little stroll through my HTTP server’s log files, checking up on varying websites I have. I found an interesting, but yet disturbing trend. The most recent visitors (over the past several hours) have all been robots! Not a single human among them! All of the users are either GoogleBot, Yahoo SLURP bot, MSNbot or MSN Media Bot. Whats up with this? Oh, and I got happy for a second when I saw an IP address I thought was human, only to discover that it was just an automated process getting a copy of my most recently blog entries (supposedly for the reading of its human though).

I know I don’t spend enough time with my network and its toys, but here is the latest revision to my network map:

Have you ever stopped to look at a barcode and try to figure out what it is trying to tell you? I never did until recently when I started using a barcode reader at home and realizing how confusing barcodes can be. The worst example would be on books. While most of my books in my personal collection have EAN encoded barcodes, which contain their ISBN number, a large number of books that my mother reads (romance novels, etc) use a UPC encoded barcode, which contains only an arbitrarily assigned number that has no relevance to anything, the same can be said for almost all barcodes on DVDs that are issued ISBNs. I have been trying to add a barcode usage functionality to GITI’s Library area, but I’m finding it hard to just use the most logical identifiers, the ISBNs. I am presently testing a pair of fields, ISBN and “UPC” to try to sort out this issue, but I find myself asking why this task has to be such a pain in the ass. When I wanted more answers I turned to the ISBN coordinating body (www.ISBN.org), and discovered that part of the reason for using UPCs for the barcodes is that they are cheaper. A person can register as a publisher and obtain 10 ISBN numbers for $30 total, but it would cost an additional $25 PER ISBN for rights to the accompanying barcode in EAN format. UPCs aren’t as expensive in bulk, and are not permenant, they can be reassigned at any time with no additional fee, so with UPC, once a book goes out of print, its number can be reassigned to a new book with no additional fee, however, it leaves the original book without a unique identifier. Many stores that sell only books reencode the ISBN into a barcode sticker for their usage, making it possible for them to opperate on a single numbering system. Should I do the same? (If you find how book stores can do this to be confusing, it is, anyone can ENCODE an ISBN into an EAN barcode, but you can only legally attach it to a product if you don’t own the rights to the EAN. To add it into the product itself [print it on the cover] it must be legally registered).

Writing code in C++ is great fun, but that fun can come to a dead halt if you realize you have no good way to compile it. So, what does one do? Well, my first attempt was to connect to a Sun Solaris system to ask it to compile for me, but for some reason CC was being a bitch and wouldn’t read my file and additionally I kept trying to call “cc” instead (cc is for C, not C++). Once getting frustrated in Solaris, I moved to trying to manually call the Visual Studio .NET VC7 compiler… apparently it doesn’t work on the systems in the labs. Next, I tried “calling home” to my Suse Linux machine via SSH, that kinda worked, it compiled and stuff, but I still couldn’t edit the file on a machine that also had access to the file to compile it, so that was a bust. Finally I gave up on the whole “remote compile” thing and went to a Solaris lab, sat down, compiled and was happy.
Since I compiled my little program I have been playing with the compilers a little more and I can now compile in the labs on Solaris, remotely on Linux and I even managed to get a command line compiler set up on my Windows Server 2003 system.
cc – Unix/Linux C compiler
CC – Unix/Linux C++ Compiler
gcc – not sure, but it doesn’t work for me.
g++ – GNU C++ compiler, works on several OSes

During the first week of my website, Clay Mentality, being online I have been monitoring its logs closely. I have watched as the expected visitors (my friends), welcome suprise visitors (search bots) and some unwelcome suprise visitors (unknown IPs, which have already started SPAMing email on the new domain).
Last week when I was preparing to unviel the site, the blog entry had been written for days, but it wasn’t until the last minute (literally, no more than 5 minutes before live) that I decided to register the ClayMentality.org domain. I was going to use a subdomain of one of my other domains for this purpose originally. ClayMentality.org is the first domain that I have registered AFTER the content was written, so it was destined to be a unique experience.
In the first day of the site, most of the visitors were friends who had recieved notification of the site via the blog. In the second day I was visited by a few friendly bots (new domain names attract attention). Throughout the last 5 days the traffic pattern has been mostly stable with return visitors (yes, you Oscar), as well as new visitors as news of the site spreads across my network of friends, and with the help of the bots, the Internet.
During the first few days of operation the site had a critical bug, the code used for protecting some of my administrative scipts from illegitimate use was accidentally being used to protect the proccessing code for the “Contact Me” form, oops. That issue is now corrected and the form is functioning normally.
The site was launched without 2 standard files in place, the robots exclusion file (for bots) and the favorite icon file; I have now implemented both, so the bots are happy and people using FireFox with too many tabs open should also be happy.
The bots that have checked the site so far include: Googlebot (Google’s crawler), SLURP (Yahoo’s bot), Internet Archive (archive.org). Websense (a bot looking for porn), and NetCraft (checking the server for its software and uptime).
So far no one has registered for the gallery, which suprises me, since I installed the new gallery system specificly for the purpose of allowing comments, rating and an overall more interacting image viewing experience.
The first week hasn’t been very eventful, but the site does feel alive, which makes me feel like throwing something 🙂

Why does Single Sign In (Single Sign On) suck? Because its centralized and one broken link kills all! I have been trying to log in to OfficeLive for the past few minutes to work on my website, but because it uses Passport to log in, Passport, OfficeLive and the interacting APIs between them must all be functioning. In this case, OfficeLive and Passport both seem to be working, but the OfficeLive link to Passport is preventing me from being able to manage my new websites through OfficeLive.
Single Sign In services are only a good idea when they work, but all people implementing them should consider having a backup plan.