Home

Well I am testing the cutting edge yet again. I have downloaded and installed IE 8 beta 1 and Firefox 3 beta 5 pre [the nightly build]. I have run both against the WASP Acid3 test and checked my website in both browsers. IE 8 scores a dismal 17/100 on ACID 3 while FF 3b5pre scores a 68/100 and when you see my site in both you can really appreciate the differences. IE 8 has weird issues with my site, which is fully XHTML 1.1 compliant for most pages, where as FF and Opera render the site nearly perfectly. I am not doing anything that I believe is unusual, certainly nothing as taxing as Acid3 but I am VERY disappointed in MS. All that money and talent and the best they can do is score a 17/100 on the Acid3 and butcher my website. What are Gates and Ballmer paying them for?
I might also mention that FF 3b5pre looks great. The icon set and the polishing really shine through. And it is quick. I can't really tell if it is faster than Opera but it definitely feels faster than IE 8. If it wasn't for the fact that some of my work requires me to use IE I might finally commit to using FF full time.

Well I started off using digest authentication but as it turns out there is no reasonable way to logout when you use that method so I have gone to form based authentication. It was not "quite" as hard as I thought it would be but it wasn't as easy as one would hope either. In the end, however, I did finally manage to beat it into submission and going forward my new access level pages should be relatively easy. I think at some point down the road I may even go SSL for the authentication and secure level, but that will require a bit more research and probably will not include a commercial SSL cert [they are expensive].

[SECOND ENTRY]

Ok, maybe I was a bit hasty on the description below. It turns out that both browsers work fine if you just set the display property back to empty [""] to make the table row visible. That is not to say that their default behaviors are otherwise correct but at least you don't have to do any browser detection and special handling to make it work the way you want it to. In addition, the same set of display values ["" | none] work fine in IE, Firefox, Safari and Opera.

[FIRST ENTRY]

I was fiddling around with adding a new secure [semi-secure] piece to the site and managed to decipher how to use the JDBCRealm with tomcat 6, it was actually pretty painless. That said the new page I was adding was to store some notes on and it is essentially a list created using the displaytag taglib with some editable parts as supplied by javascript. I discovered, much to my chagrin, that IE and Firefox differ somewhat in the way they handle table rows in that in IE you can make a table row visible and invisible using the css property display: [block | none] but in firefox you must use display: [table-row | none] or the table will get progressively out of whack even though the in memory html "appears" to be correct. I find Firefox's behavior for this worse than I find IE's. I agree the W3C page indicates Firefox is doimg the right thing by using the none and table-row values but at least when IE doesn't understand the table-row property it doesn't do anything. I find it preferable to have it do nothing [IE] than to see the table's rendering go nuts when the diplay property is wrong and wonder what the heck is wrong [Firefox]. Don't get me wrong, I think both browsers are good and valuable, but I think that, in this instance, Firefox would be better off ignoring the block value on a table-row than doing what it does.

May it be a safe and joyous. And in case you are wondering, Phil saw his shadow, so six more weeks of winter. You can view a video [wmv format 47.6MB] of his prediction and visit the websites at visitpa and groundhog.org

I was reading about a centering issue that I was having with this page in firefox and one of the posters said that text-align: center is not supposed to work with block elements only inline elements within a block. I find this kind of thinking a bit simplistic. While I might agree that the specification states this is the case I believe behavior of nested block elements should be to inherit the same characteristics as inline elements. This feels more intuitive. Yes IE is not strictly compliant in this regard but having to actually state that an element should have auto margins when the parent has already set a "text-align: center" just doesn't make sense. for example if I have
<div style="text-align: center;"><div>abc</div><div>
I expect abc to be centered in the page. Simple inheritance should ensure this, but apparently the CSS spec seems to indicate that the inner div should not inherit the properties of the outer div. This seems not only counter-intuitive but stupid. It complicates the ability of the page author to depend on inheritance to style a page. And to be honest I just don't like it.

I will acknowledge that the inner div may indeed have the text centered within it but if I expect the div to have the text centered within it then I expect the div itself to be centered as well. If I want different behavior then I should have to specify it then, not have different behavior and have to specify the behavior I would have expected. That sounds convoluted but it is an expected, intuitive behavior.

This is just too funny not to share middle class windows

That was not my only change for today. I was testing how well the site works with google using the site keyword

e.g. "search phrase" site:sol2charon.net

and came across google's webmaster pages that actually give you feedback on things like broken links and site ranking as well as cacheing, links from other sites and more. Anyway, what I realized is that I did have some broken links so I did a little more clean up in my Articles and Information sections. If you are looking for more info a google search for webmaster will work pretty well but here is a direct link to Google Webmaster Central as well.

OK, I may have been a bit hasty with the Nokia PC Suite software. It does work if you disable all of the security software running on your computer before you install it. I have seldom if ever needed to do that before but it may have smething to do with the way Vista handles USB which may also account for the problems with the APC Software. It took nearly 2 hours and a phone call to Nokia USA Support to resolve the connectivity for the phone so I think I will wait a little longer to try to correct the APC issue.

New year, new entries. I hope all of your holidays were good. Not really much to report at this point my self education has slowed a bit but things are evolving at work so that I may have some new things to discuss here soon. I did do some interesting work on some authentication that may develop into something to offer some real insights on but for now I think I will refrain from making too many comments on it until it gets a more thorough testing.



I have some career advice to offer anyone reading this since a friend of mine has been discussing it with me lately. First, if you are unhappy with your current job it is your responsibility to act on that dissatisfaction, whether that means talking to your boss/chain-of-command or putting out your resume and approaching head hunters/placement centers is a function of how unhappy you are and how well your personal career goals fit with your current employer. Second, asking for something from your employer is almost never a negative, whether that is time off, a raise, a change of position/title, feedback, etc. the worst that can typically happen is that they say no and the last I knew, no one ever died from being told no.

Lastly, very little, if anything is ever resolved by just complaining about it, here are two additional tips: 1) If you have a complaint don't say it out loud until you can provide a solution to the problem you've identified, complaining just for the sake of complaining has a tendency to bring you down and anyone you are talking/complaining to. 2) This is always true, if you spend the time you would spend complaining about something actually working on a solution/resolving whatever it is you would complain about, whatever it is you have to complain about will usually be over and done with in far less time. Thus the old expression "If you did what you were told when you were told you'd be done with it in far less time than you spend complaining about it and you'd have nothing left to complain about."



A few more Vista insights: APC power chute exhibits some unpleasant behavior in Vista 64 bit, it's not a knock on the actual UPS, just APC's programmers. Nokia's PC Suite Software does not recognize my 6263 phone on Vista 64 bit even though the software says it is Vista compatible, the OS recognizes the phone when it is connected, and the manual indicates that the phone is compatible with Nokia PC Suite Software. Oh yeah, and just where is IE 8, since apparently the cryptographic mimetype handlers in IE 7 do not work correctly. I recently had to renew my Thawte Email Certificate and ended up having to do it on an old XP SP2 laptop then export the key to my Vista machine to install it. And before you try to tell me Firefox would have worked, I tried it and it didn't [both 3.0b2 and 2.0.0.11]. I guess, I should add that I am not surprised since IE can't correctly render xhtml when it has the xml declaration at the beginning, most likely due to its mime type handler as well.



An interesting style attribute "padding-left", it is providing me with some indenting using an empty span element. Something to keep in mind should you ever need/want to indent in html. Happy coding.