There and Back Again

I’m in Chicago its friken cold

Megan and drove here and took pictures and notes of our journey, i’ll try to get them up eventually or at least link to them on her blog

Oh and thanks to weather.com for the image

Blog spam has started to find its way into my moderation queue again. To help cut down on the amount of junk I have to delete by hand I added ringtone to my kill list. So if you want to make sure your comment is never seen, toss ringtone in it and it will be gone.

I got tired with the look of my blog so I updated it. I’m not 100% sure i like all the changes Mainly with how the lists on each side of the content turned out.

The site is optimized for higher res monitors and doesn’t work without horizontal scroll at anything less then 1024×768, but hey geeks read this site, they should have some resolution. I guess I should be concerned that this pushed my Google ads off those same none wide screens, but I guess thats the tradeoff for more room for content.

Anyhow tell me what you think, biggest thing im worried about right now is the light gray background the main content area is one. All that white seemed like too much but now the gray seems worse.

And yes the header is 1400px wide because thats the width of my laptop screen and I didn’t want to look at it running out while I was reading peoples comments.

Update:
I fixed a number of things people pointed out. I ended up cutting the width of the sidebar by 5px and the main content area by 40px. This gives you all the main navigation and content at 800px wide. You’ll still have to horizontal scroll to see the archive/category links, but I like this better then the huge single bar especially since that part of the nav isn’t that important. I also shrunk the textarea to fix in the content box, and moved the ads below the navigation on the right. I can also play around with them if the new placement means I get fewer clicks.

Since the begining of the project i’ve been subscribed to the fedora devel list. I read it too see whats in the new releases and to make sure I know all the tricks when im upgrading my server. I don’t post often but its a pretty good way to get an idea of whats happening in my distro of choice.

Over the last week a thread has been going on that has me convinced I know the reason Linux isn’t ready for the desktop. Many people just don’t get what developing an operating system is about, I’m not even sure they get what using a computer is a about. And even if they aren’t a majority they slow down the approach to sanity so much thats its taken Gnome since 1997 to get to windows 95 levels of usability (and hardware integration isn’t even to that level yet).

To sum up the thread, Gnome is removing the open terminal item from the right click menu of Nautilus, and a number of people think thats such a big deal that Fedora should make an exception from its policy of using upstream code whenever possible.

The reasons they give are odd in a number of ways.
The terminal is the most important part of unix (if your a graybeard maybe)
The right click menu is the easiest way to open it (kinda odd since a terminal junky would always say a keyboard shortcut is better)
The terminal should be used by everyone since its so much more efficient then a gui
That having to use a terminal to get something done isn’t a bug

Now whats also odd, is that there is already a package to give you this open terminal functionality back if your really want it only its better since if your in a window in Nautilus it open in that directory.

So whats my problem with these arguments.
Wake up, Fedora and Gnome are about being cutting edge, how is pushing the terminal as part of normal usage part of that.
The pointless rightclick argument, its already in the menu and you can drag it to the toolbar for a launcher
The terminal being more effiecient might be true, but thats only after you’ve managed the learning curve.
To change to a directory in nautilus move a file all I have to do is point and click, then drag and drop, its a pretty basic skill that everyone knows and since it has good visual feedback its reasonable to figure out.
To do the same thing in a terminal I have know the following cryptic commands (they are meaningless just because you and I have long memorized them doesn’t mean there not)
cd, ls, mv

I’m not sure why someone who uses the computer for surfing the net, checking email, and writing papers would want to learn these commands, but a very vocal group of people on fedora devel thing were doing them a disservice by not making them memorize them. Many even seem to think everyone should learn bash scripting and awk and crap I don’t even have a need to learn in order to be good computer users.

The reality is most people don’t want to become computer geeks.
A computer is a tool, just like there toaster is a tool.
It should work well, most of the time, and not be too difficult to figure out.

Thats what people want out of computers and thats why they use Windows or Mac OSX, neither of these are perfect but at least there developers have the right idea.

So I had a linksys wireless router that I use to share internet between my laptop and megans ibook, its an older 802.11b job a befw11s4 or something like that. So Megan got a new ibook today, yeah I know the new intel mac’s ones are coming out but thats not for a year and the current 800mhz thing is slow and was the old OS and had a 12 inch screen with a poor viewing angle. So we got a new one, nothing insanely fancy but its a 1.3ghz with 786meg of ram, and to make a long story short it broke the router.

So you connect to the router and about a minute latter internet stops working, reboot the router and sometimes the cable modem too and things work again. Very odd stuff, so I just went out and got a new one, one of those wrt54g models that run linux. Once I ignored there setup program that didn’t work right and just used the web interface everything was good. I now have working wireless, and its even g on the ibook, of course I somehow managed to get a new laptop with only built in b.

Some of you may have noticed Googles newest project, Google Web Accelerator. Its some sort of combination of web proxy/compression engine.

You might also have seen the Slashdot article about people having problems with it. Now I know i’ll never use it (I’ll keep my surfing habits private thank you), but if the problems linked from slashdot are true, everyone who makes web apps should be concerned.

It seems that google is randomly prefetching pages, and caching output as much as possible (more then they should?) and sharing it between users.

I’m not sure what the next step is, the gwa FAQ seems to imply that the web master can control what is prefetched, but the articles imply that google just fetches like mad.

Anyhow I’d be greatful if someone has found a report of what gwa actually does to post a link here, so I can decide if gwa needs to be blocked to keep it from messing stuff up.

Update:
After looking at the GWA feedback group I’ve come to the conclusion that its really doing some odd stuff with cookies which seems to break many features on forums apps like phpbb. It also seems to randomly prefetch on sites after you’ve logged in. So for the time being the only safe thing to do is tell everyone you know not to use GWA. If someone is running a forum you might want to add a blocking script.

Everyone was filling this out on Planet Gnome, so I thought i’d see what my profile is. Just as I thought I speak nice boring general american english, though I think before I moved to Arizona I spoke a bit more Upper Midwestern.

In November I came back early from my time in London to join a small Open Source php company here in Phoenix. I wasn’t sure if I was going to stay in Phoenix long term but its managed to work out that way, last week a sign a lease for an apartment at Indigo Palms.

Anyhow, work has been extremly busy making it hard to do anything at all on phpDocumentor of PHPca, but there is a bright side to that. Were an open source company and were slowly releasing our projects out as open source. The big thing im hoping to get out soon is our Celini application base, so I can talk about some of the cool stuff were doing with JPSpan.

On that note I have a small proof of concept demo of what can be done with direct javascript rpc to your php code.

The context is an editor for fee schedules in a medical clinic, there are something like 18K codes in the full table, I only have 1K loaded in the demo. You can scroll up and down (just links any information on using scroll bars in js for this would be appreciated) with information being loaded as needed. You can also click on the fields with a border to edit them, click on another cell saves the first one.

There is still a bit of work to do, mainly in filtering, jumping to any row, better edit interaction (enter saves, keys bound for up, down), and maybe a scroll bar.

Anyhow I thought I’d let the world have a small peak at what im doing these days, and hopefully i’ll have time to get the source ready to share in a week or too.

View the demo (NOTE, this code will eventually be released as GPL).

I played around a bit more with my blog, adding in links to other sites and starting a blog roll. I also imported all my old photos into gallery (which i learned to hate more) so I don’t have a reason to keep the old website around.

So now http://JoshuaEichorn.com redirect to this blog, all the old links are still good i just moved the index page out and setup a redirect.

My gallery photos are linked on the links section on the right now, out of the old stuff that I imported, think the best are the photos from the Hoover Dam, and From a walk I took up to “A” Mountain after a Storm.

Oh and I somehow lost the images I had in the root of my blog photos directory, I think the only thing I ended up losing was a photo of any annoying keyboard so I guess i’ll manage.

Interesting sidenote, I looked through google images to see if i could find it there, but no luck, they had pretty much every other image, it must have been deleted awhile ago and I just didn’t notice it.

So I want through and cleaned out the worst of my referrer spam today. Its just fixed this month, since webalizer doesn’t rebuild a page when stats change.

I just used the HideReferrer config option with some * rules to get rid of my mortage spammers. I need a better long term solution, I think for starters im going to create a php script that is prepended on all my webalizer files and rewrite the referrer section with the nofollow attribute.

The next step after that is too look for a new stats package. Anybody got any suggestions.

I’m at the Scale Linux conference in Los Angeles today. Its been ok, was expecting to meet more of our customers. My boss David is speaking next which should be ok.

Oh and my sister found a bunch of old photos I took and got a photo cd made. Does anyone know if gallery has a cli script for doing batch uploads.

Update: You can view David’s slides at his blog.

Even if you have half a pinky you can hit this enter key. Say goodbye to using | or \, and now I can’t reach the delete key as easy.

I’m safely back in Phoenix, 24 hours of traveling fun. I’ll post more details when im not so tired.

Ok so I got more pictures uploaded from my holiday in Europe. I got lazy and stopped renaming them so you’ll have to what the # ones are pictures of.

Oh and for all of you that said gallery is great, your insane. Its got a horrible interface, I constantly have to change the same setting over and over. It has limited image sizes.

Of course it does have the spiffy XP upload wizard, so until someone writes some new gallery software that Just Works. I’m stuck with this.

Oh and go view the photos some are actually good.

I got all the pictures from Rome renamed and they are uploading as I type this. The Panoramic shots will be a bit, looks like im going to need to find a better program to stitch them together, or at least i’ll have to try the other slower program I have.

Anyhow for your viewing pleasure, 101 pictures of Roma.

I added a firefox logo to the site from Spread Firefox, it looks to be a good attempt to manage grass roots marketing.

I also wrote a little hackish plugin to allow me to display info from my profile in my template. The result is the the Author box on the left. You can download the plugin from my Worldpress Plugins page.

Update:
I also did some css tweaking, the calendar should suck a little bit less, and I turned off the Opacity hacks for now.

I’m back in London and Sue and Megan and on their plane.

I had a good time on the trip, saw lots of neat stuff, got annoyed with my fellow travelers (they’re women after all), spent too much time on trains, and saw lots of old Stuff.

I’ll be attempting to get back into the work swing.

But on a more exciting note, pictures from the trip are getting uploaded into my gallery, and Megan is back dating blog entries for the trip.

Im in Rome things are going well. Still left are Florence, Prauge, and Amsterdam.

I’ll be traveling around Europe for the next couple weeks. I’ll have a couple hundred more pictures when I get back.

I should be checking email again on the 21st or the 22nd.

Php London was tonight, great group. We spend less time presenting then at azPHP and more time drinking but there is nothing wrong with that.

We had some very good converstations on templates tonight.

Templates we talked about:

• Smarty - No good things to say here (fat and slow)
• Flexy - Was the prefered solution of some, compiler with light weight runtime
• Savant - My prefered solution but I might not be the normal use case

I just made the last train home, and im a little pissed, so its not the time to go into details, but what makes a good template engine is a topic I would like to discuss more.

And the article by Harry that started it all: Templates and Template Engines or why do we need to make up a new language when we have PHP