It’s true - one man’s junk is another man’s treasure. I know this to be true because I was the recent recipient of such a fortune. Back in October, I was talking to an old friend of mine over instant messenger, and I had mentioned that I was working with some old UltraSparc1 machines (thank you Ed Corrado). My friend thought I was wasting my time, and mentioned that he was about to throw away 4 old Dell workstations and an ancient Dell server. A quick arrangement was made, and I set a date to drive up to Manhattan and pick them up. (more…)

I was originally going to title this post “The Reaon I’m a Terrible Systems Administrator”, but on further reflection I changed it to “The Primary Reason… ” and finally to “One of the Many Reasons…” before I decided to give up and put the title more in context. For now I’ll just thank my lucky stars that my current gig has me doing heavy PL/SQL development instead.

I’ve wasted a lot of hours the last two weeks trying to troubleshoot issues with my network at home. In two specific cases, I could have avoided a lot of frustration and grief if I had only checked the log files immediately, instead of pounding google and attempting a hodge-podge of fixes that didn’t work (more…)

gnujersey.org has a new look. Dave Harding provided the requirements, badges, and feedback. I hacked the css, added the required tags, and made the site look pretty.

There were requirements besides an updated look and feel. Standards compliance and browser compatibility were important. lynx(1) (a legendary text-mode web browser) support was specifically requested. This was easy to achieve because the original code was in great shape. Attractive printer output was Dave’s last requirement. I achieved this with a special print stylesheet that formats the page in black and white for a letter sized sheet of paper. You can see it in action by clicking on print preview. Dave will be distributing printed copies of gnujersey.org at computer festivals and other events.

I always have trouble selecting an appropriate color palatte - but in this case I got a pretty good idea from the start. I latched onto the theme that Jersey is the Garden State and worked from there. I’m kind of caught up in fall, which is one of my favorite seasons, and I’m impatiently waiting for those leaves to change colors. I decided to play off that with the orange link colors and the calendar icon. And you can’t have the Garden State and not have any green, so I added that to the logo and to some of the bullets on the sidebar as well.

Dave and I created each of the graphics using the GIMP or Inkscape. Both of us hacked using vi clones; Dave used nvi, and I used vim.

It’s a very simple design, and I think it serves the site well. If you have questions or comments, let Dave or me know, or post comments to this entry.

I personally like fixed width web pages - specifically ones that fit in a 800px screen without horizontal scroll, but according to Jakob Nielsen’s readers it’s the number 9 on the Top Ten Web Mistakes of 2005.

Last time I really paid attention, this was all the rage in web design, and I did the same thing with my site back in the spring when I updated it. So I just went out to do a survey, determined to get examples from the top web designers to refute this. Unfortunately, I came back a bit empty handed. There are still a few hold-outs - Zeldman, Todd Dominey, and Mike Davidson among others. There was an argument (sorry, I forget where it was posted) that users have difficulty focusing on text contained in too wide an area, that it’s hard to visually scroll to the next line, or get back to your spot if you’re momentarily distracted. The traditional layout of newsprint was cited as an example. But I noticed that in the last round of redesigns, most people are adopting designs that flow into the window size. Most of these are still optimized, or at least still fairly readable, at 800px wide, but gratiously widen in a larger window.

I personally like 800px width because I have my window resolution set at 1600×1200, and it allows me to have the web browser open in exactly half the window while I do what-ever else it is I’m working on in the other half of the window. Surely buying another monitor could resolve this, but I’ve got enough desk clutter to deal with.

One of the primary points that Nielsen points out in this piece recurs from a web design mistake much higher in the list… fixed text sizes. Given my resolution preferences, sites that have fonts fixed down to 8pts are painfully hard to read. I have to use the old firefox ctrl++ trick to bump up the font to a readable size. And this is where a flowing layout is really important. If you have your layout firmly strapped down and then someone blows up the font, they are going to see some strange issues where the text no longer fits inside it’s container. For an example, just blow up the font (either using browser default, or by hitting ctrl++) on this site and see what happens to some of the headings to the right. Yech… that’s ugly.

So rest assured, in my next re-design I will be addressing this issue. The site will still be created so that it is perfectly viewable with a decent font size in an 800px window, but it will flow gratiously if you have a nice wide screen monitor.

Aside from the fixed vs. flowing debate, there is one thing that really aggravates me and that I think is almost a laughable oversite by many designers. They’ll build a site intending there to be no horizontal scroll at 800px, but they forget to take into account that the veritcal scrollbar takes up a small amount of space. The result is usally 1-2 characters cut off on the right of the screen, and the tiniest bit of horizontal scroll. You really need to make your site a little narrower than that. I generally use 770px. And when designing sites to be easily printed, or designing your print.css, be sure to be kind to the rest of the world that uses A4 instead of letter sized paper.

Been a whole season since I blogged
Whoah… what happened to summer. Last time I blogged I was on vacation, and now I feel like I definitely need another one. Things around here have been fairly crazy.

Beer Brewing Continues
A few batches have come and gone since May - some good - some not so good. In June we sampled our Blackberry Wheat and found it was good. We brewed a brown ale. In July, my brother carted the Brown and the Wheat up to MI to our parent’s, where several family members and friends of the family saw that it was good, and helped us drink a good portion of it. In late July we made a Pilsner - an attempted 10-gallon batch that we only got about 7 gallons out of. The verdict is still out on that one - it needs to lager a bit longer. A few weeks later I made an Irish Red with my friend Jason. That one came out really well (I still need to send a bunch of Jason - I haven’t forgotten J). The week after that, my brother and I made a brown ale. Something went wrong with that one. The yeast never settlted out of it. It’s got some funky off-flavors. I’m hoping that it will eventually age and become drinkable, but it doesn’t look good so far. We’re getting ready to do a clone of Magic Hat #9 this weekend.

Kitchen Cabinets are DONE
After a long battle consuming several years, I’ve finally managed to get all of my kitchen cabinet doors stripped, repainted, and hung. What a mess that’s been. I’m thoroughly glad it’s over.

Bathroom Renovations
I’m in the middle of fixing up the bathroom. Of course this has been like any project around this house… what starts out as a simple project - I’m just going to repaint and put a new medicine cabinet in - has become a major undertaking. I’m going to put in a larger medicine cabinet, and some proper GFI outlets, which means I gotta make a bigger hole in the wall, move some studs, run some electrical cable, etc. And last night I almost killed myself cleaning - a heavy concentration of TSP and bleach does wonders against mold and mildew, but in a small room such as a bathroom, it can be deadly - even with the door and the window open and the fan blowing.

Survey’s Up
I completed my first major freelance coding project recently. This one took a lot out of me. In addition to my regular job, I put 80 hours of work into it. It was a fun project, and I learned a lot of neat tricks in PHP/mySQL and generally advanced myself in knowledge. With all my previous PHP/mySQL work for lunabyte systems and working with (hacking at) several other scripts, I felt I had a fairly good grasp of the language/environment, and was probably an avarage coder. I think I’m advancing past that point gradually, and my bag of tricks is groing substantially. Unfortunately, this is starting to seem like the project that will never end, because after two weeks the clients are still finding all manor of changes and modifications that need to be done. I really noticed that it’s almost impossible to do a project right without having the time to schedule meetings with clients, properly review requirements and specifications, etc.