Just found a a contest site with a chance to win $1000 cash. They are calling it Project No Name. The site says that two companies are coming together and are wanting a new name. So they are holding a contest, and if they select your name, you win .. that simple.

It's pretty cool to see companies using the internet for things like this. I had to Dig it http://digg.com/design/Project_No_Name_Wants_You !

Lately I've been getting a ton of spam on my MSN Instant Messenger account. It is quit annoying. I could of just modified my privacy settings in my client (on pidgin Tools .. Privacy, select MSN and block only users on buddy list). But I do a lot of work with open source and communities, and would like to have people who need to contact , actually be able to contact me.

I found a few solutions. First was logging into account.live.com and modify my shared profile. Under social, I changed the permissions to just me. Hopefully that helps a bit. My news step will be using a pidgin plug-in called Bot Sentry. Which will ask users not in your buddy list or approved list to answer simple questions before the message goes through. Ex. Please spell out the number 5. So it should stop all that automated spam.

If I have to go that route, then I will be sure and post a follow up.

Just got down spending about 3 hours trying to get a simple gallery page to render the same across multiple browsers. It is crazy trying to get all browsers to work. I think we are dropping IE6 and FF2 from strict compatibility tests. As long as the data is rendered, we are calling that good. But that still leaves, FF3, IE7, Opera, Safari and Chrome. And we have IE8 around the corner, which will initialy just be a render data in a readable format.

The issue we are having is inside Webkit based browsers (Safari, Chrome, Midori, and many others). The have our CMS applying a wrapper to a basic php gallery page. This page uses jquery and galleria to handle the image transitions and such using ajax. Works great on the browsers by itself, and after a bit of CSS work for IE7 works great on that and FF3. But once done, Safari and Chrom wouldn't display the ajax content when the page is loaded directly. Hitting refresh though, the ajax / javascript engine inside these browsers decides to work, and everything works great. It is the craziest shit I've ever seen. Nothing changing besides hitting refresh.

Which brings on to my biggest compliant, UI design sucks. It isn't even for Web Design, but designing applications to work on a variety of interfaces is going to cause problems. Windows has done a decent job of not changing things too much between major backends, so apps developed for 2000 should look find in XP and Vista ... should, but not always. But look at java apps, that share the same code for Mac OS X, Vista, XP, Linux (KDE, Gnome, Fluxbox, etc, etc) and it has some small issues that araise, you got font issues, DPI, resolution, window decorations (or the lack there of). It is a huge issue. And people (end users) don't look toward the person resonable for their platform (firefox, IE, Apple, Gnome) but instead to the individual developer to create special cases for all these different idiosyncrocies. And we, as the small developers, bend over, take it, and try to get all the wrinkles ironed out.

While choice is good for the consumer, it sucks for those trying to create things to work for the most amount of people. Next to writing a custom interface for every single option out there, there are going to be some that run in a degraded mode, ignored, and in the extremem cases not even work.

Okay Rant Over - back to work.

In a previous post about Redmine I mentioned that I was having problems with LDAP and 2003 Active Directory authentication. Now I considered myself to have a fair amount of LDAP and Active Directory experience. I have set up samba PD and set up PAM to auth against LDAP. Even to grab user credentials. I've written scripts to pull users from ldap for initial setups in other applications, etc, etc. But for the life of me I couldn't get redmine to auth against AD.

I probalby spent about 5 hours debugging Redmien to find the error. I had logger.info lines all over the place (I'm a ruby newbie), I'm debugging the mysql queries, if statements, everything I can think of. After 1) understanding Redmine and 2) understanding ruby, I was able to start norrowing it down. The SQL queries were fine, the initial ldap bind was fine, but it was authenticating.

I ended up finding out that the problem was with the ldap_conn.search() fucntion. I wasn't executing correcting. They seemed to using it as kind of an if statement, so I was trying to ouput the status, but wasn't getting anything. I thought this was my lack of knowledge of ruby. I forced the ldap_bind to fail, so I know for a fact the ldap_conn.search would fail, and it did, but this time it would return false. But when everything else seemed to work find, ldap_conn.search would return nothing. I was stumped. I had no clue what to do. I went home that day feeling defeated. I talked to Brian and he made fun of me, like any good friend should, and told me it was user error. I laughed him off and went on my. Next day at work though ended up proving him right. I ended gettign the bright idea to cdump out all the variables being used. I read that when the ldap search function was giving a filter string that was the written out the way it wants, it silently fails. I looked through them all, and they all looked okay, fiddling with them for a bit didn't do anything. Then I dumped them again , and it hit me, like a pallet of falling bricks .. yes it hurt as bad also.

My username field for the ldap search as set to sAMAccountName , what you can't tell from this is that it was actually 'sAMAccountName ', with a space at the end. So the filter ws being set to sAMAccountName =ryan.gibbons. While you would think this might work, it doesn't, but whats worse is that redmine let this pass, and ruby-net-ldap didn't doesn't handle invalid filters at all. Hence my post below about Data Cleansing.

So hopefully this post will help others out there having similar issues. I've learned my lesson for the week, only question is how long will I remember it. Tomorrow I will go submit a bug to hopefully have this error fixed. Yes even computer nerds and programs fall victim to PEBKAC.

Should be enough said, but apperently it isn't. Yes, I had a dumb user moment, and had a space at the end of one of my inputs. Which the application saved, and passed on to another "plugin" and failed to make the connection, silently btw. The space should of been cleansed out in multiple places, especially a trailing space.

I am pretty sure I learned this in my first year of school, programming 101.

Reminds of a comic .. from xkcd

I've started to use Redmine a bit in my personal business to help manage the few projects that go through my doors. A buddy, Brian Wigginton, turned me on to it. We have spent several nights looking for a good app that can manage, tasks, issue and project workflow from beginning to completion. After trying out several different ones ( i.e. OPProject, Trac, Achievo, and even dokuwiki ) based on multiple technologies ( i.e. java, python, php, and ruby). We ended up with Redmine, which is based on Ruby on Rails. The install wasn't too bad (we spent most of the time working with phusion passenger, which being a fan boy of mod_fcgid, I highly recommend.) The database design seems pretty straight forward and it has all the features we were looking for.

OnePoint Project (OPProject) was an amazing it. It was java based and we deployed it over a tomcat-6 server. For a strict project management app, I think this one wins hands down. The interface is clean and sharp. Even know little about Project Management, I was able to find my way around the system and get some basic things set up. It manages resources and time. And again, the Gnatt charts were very easy on the eyes.

Trac is more of a software management portal. It is based on python, and I deployed it in the past using mod_python (before I knew about mod_fcgid). It is a very strong app for managing software development. It has a strong integration between the repository, bug tickets, and a wiki. It was missing a few of the project managemenent aspects we were looking for, but it works great for basic management of software development.

Achievo describes itself as a flexible web-based resource management tool for business environments. And I would have to agree with that. It doesn't tie into a software repository like Trac does, but it does provide more project management tools such as gnatt charts. One piece I did like, was it's ability to add clients, and manage projects and task for assigned to those clients. If we were looking for a general business or department management tool, this peice of software would of probably won out.

DokuWiki is just a wiki, with the ability to add plugins. We started using it in the interum as we were trying to find a software package for us to use. You could use this as a basic project management for a group of people, especially a decentralized group. Collabrating on pages for ideas, managing picture, and providing revisions is essentional. There wasn't any strong project management tools, but getting going, it was extremely usefull. I love the fact that it doesn't use a database for it's wiki pages and revisions, but instead plain text. This has always been my love of dokuwiki, and will always make it my first choice for small to medium sized wiki's.

Redmine is really a plethora of applications rolled into one. It provides Repository integeration, bug ticketing, wiki, forums, customer management (via plugin), and project management tools. We are able to track our time for each project, and with a plugin, see a resources time across all projects. We have a wiki, document section, and file upload for each project. Wiki pages can cross link to other projects, and any one section can refernce another. I.E. a wiki page can reference both a ticket number and a repo revision. We are also able to provide some basic project planning through gnatt charts as some of our customers request. Security is strong, and development still seems very active.

We've been happy with our initial testing of Redmine, and look forward to really running it thorugh it paces on our next project. I've also started installing Redmine at my office as management is now starting to ask for project schedules on some of my long term projects .. woohoo :/. With our personal install, we were able to get LDAP Authentication working with our Zimbra server. At my office, I am trying to get it working with Active Directory. It seems to be working, but is failing on the search. Debugging Ruby is something I am new at .. which is a topic for another post another day!

So most people are talking about the rise of Firefox, and the market share it is capturing. But one thing people are failing to mention is the market that these stats are taking from. If I take one of my own communities (granted very small) which focuses on open source on laptops then you would find Linux OS 42% and Windows at 55% (amazing how even an Linux driven content still has a majority of Windows OS as it user hits in the past month). You will find Firefox at 61%, IE at 22%, Opera at 7% and Mozilla at 6.5%. While the OS scale isn't what I expected, the browser market is about right (especially when considering the inflated Windows OS numbers).

Now, that was a tech market, lets move to the business market. Again I work for a Project Management training and consulting firm. I am registering Windows at 96%, Macs at 3% and Linux at .5%. This is about on target with other publishing authorities. For the regard, other OS'es registered are SunOS, PalmOS (my cell :/), iPhone and SymbianOS (the CEO's cell). Pretty interesting. Now, for the kicker the browser market share. While seeing IE at 77%, Mozilla at 20%, Safari at 2% and Opera at 1% doesn't shock me too much, what does it the versioning within the two biggest, IE and Firefox.

The majority of our Internet users are using IE 6.0, making up 50% of IE traffic. That is nearly 40% of our site's visitors. Just under the other 50% is IE 7.0, which the other .5 percent being IE 5 and IE 8. Come on, are you kidding me. Their are still companies that haven't moved to IE 7 yet. I understand not doing it the first year. But seriously. IE 7 makes a world of difference, for both the end user and the web developers. In my eyes, this is just unacceptable. If you have a corporate web site or app that requires IE 6, then you need to get on the move and get it updated to dump IE 6.

Firefox is moving along nicely. Considering it has only been released for a few weeks now. We are registering 30% of our users running Firefox 3.0. I am showing just over 1% using 1.5, with the rest of the FF users running some version 2.0. This is about what I expected. I am expecting by end of year to have FF 3.0 be over 75%. Unless they push FF 3.0 updates over their automatic update, then I expect that number to be closer to 90%. But as far as a web programming perspective, this isn't as big as a deal, because FF 2.0 follows the specs set forth by HTML and CSS pretty closely, as does IE 7. The reasons for end users to upgrade remain the same, better security.

The last spec I thought was interesting, and have seen reported elsewhere is screen resolution and colors, these are very important. 98% of our users use a resolution of 1024x768 or higher, which 35% using exactly 1024x768, only 2% use 800x600. Which is great when desining a site, the extra width planning for 1024 verus 600 is awesome. Screen colors actually suprised me a bit. I would expect 32bit color to be over 95%, but it isn't. it is just under 90%. 24 bit at 3% and 16 bit at 6%. Interesting.

I am not going to do any grand finale conclusion, as those are mostly crocs of shit anyways. What I am going to leave you with is the numbers and let you draw your on conclusions. These numbers don't tell you how much better any one product is compared to another or better adoption of one to the other. Just the stats over a month.

I'll admit it, I was a bit nervous about using b2evolution. The admin interface in wordpress had set my standards pretty high. But, much like my distro of choice, I like to choose things away from the norm. I can report that I have been pleasantly surprised.

First off the admin interface is very sharp. The new 'chicago' skin adds a nice tool bar type interface and a pretty good layout of everything else. While it doesn't have some of the web2.0 features that wordpress has, that is okay. I don't care that much about eye-candy in my admin interface, I just want a clean look and something that works. One interesting things to note here, is when create post, there aren't tags like most modern blogging tools provides, but there are categories. You can choose your main category, and then add others, so in a sense, this provides your tagging.

Another neat feature, or at least it seems to be b/c I haven't dived into it too much yet is the Type of post. By default everything is a "post" But you can change the type to Link, Page or Podcast by default install. You can also define your own post types, I am not sure what all this defines, but it sounds cool, and it sounds like it would display them diffently. As for static pages, in my menu bar, I have added the plugin for to display pages, and then set the post type to page. It then shows that post as a page instead of in my blog ... a different school of thought, but so far it makes sense and I like it

The other nice feature seems to be it's skinning. While it doesn't use a default skinning engine, it seems to be a pretty simple one. They provide default layouts for all the sections, so you can just copy and paste it, or if you don't need to modify it, you can just include it in yours. The code seems to be pretty explanatory, and without a lot of fluff. Right now I am using one provided by b2evolution install, but I think I might look into convert a CSS theme over to b2evolution in the near future

Overall I can say I am very happy, and I am still just scratching the surface as to what it can offer. With multi-user and multi-views, the possibilities are really endless. (Okay not quite endless, but close to it)

Well, I am now a Class B no-www site.

This means that I politely and quitely redirect all incoming www.gibbonsr.net to gibbonsr.net. Why? B/C it is a waste of time to type it and tell it to someone. Back in the day, I might see why this is necessary, and it some cases it might still be (where you want a different site under www.domain.com versus domain.com), But these cases are very rare. Most of the time, the redirect without www to the one with. A few sites don't do either (digg.com and flickr.com).

Even though the site is active anymore, the content is still valid http://no-www.com