Categories
Technology

Browser testing

Roger Johansson details his browser testing strategy and it is far more extensive than mine–at least for CSS and markup, though I go much further when it comes to JavaScript.

  • I start with, and use extensively, Firefox on the Mac. The main reason why is the extensions, specifically Firebug. I think that Joe Hewitt’s Firebug is the third single-most important component of today’s new web development–following on Mozilla’s innovative architecture, which enables such extensions, and REST.
  • I then test with IE7 on Windows XP. Why? Because if anything I do is going to break, it will break with IE. I no longer have an IE6 box, but I do use Total Validator to take screenshots in IE6 and Konquerer if I’m working on issues of design.
  • Next, I test with Opera on the Mac, which helps me discover those things that Firefox allows that aren’t necessarily standard. I find Opera to be the most standards compliant browser.
  • Then I go to Safari and the most current WebKit, both still on the Mac.
  • I need to test with Camino and Flock more. However, my logs tell me I have people using IE, Firefox, Safari, some Opera, rarely Konquerer, and older versions of IE, on Windows and the Mac. These are my target audiences.
  • I tested the book with OmniWeb, but I don’t have it any longer as it’s a ‘cost’ browser and the cost isn’t justified.
  • I test with Netscape and Opera on Windows XP, last. I used to test for IE6, and I did for the book. However, I don’t have access to a IE6 machine, now, so am dependent on IE6 users to tell me if something breaks.
  • I provide a mobile stylesheet, which Ralph at There is no Cat, was kind enough to test for me. I also use Opera’s mobile feature to test.

One thing I talked about in the upcoming Adding Ajax book is understanding your audience before making a choice of target test browsers. If we use progressive enhancement as a development approach, which means creating the functionality without the use of scripting and then gradually adding script effects, then we always have a natural fall back if a script effect doesn’t work with one browser or another: just disable the effect for the browsers that choke. Those few who are still using IE 5.5 on the Mac (why?), or IE 3.2 (WHY!), at least get a decent shot at a workable page, if not a terribly interactive or visually appealing page.

Then there’s my feeds. If all else fails, I provide full content feeds.

Roger’s mention of semantic markup is spot on, also. I haven’t been pursuing this as diligently as I should, and plan to go over my design one more time and look for better uses of markup—after the book has gone to production, of course.

My biggest design problem? Fonts. I can never find a font that seems to look good everywhere, and that scales as well as I’d like for each resolution. That’s mainly because I haven’t taken the time with fonts as I should. Another thing to explore as soon as the book is finished.

Oh, and this site is using a conditional IE stylesheet.

Categories
Burningbird Technology

The wonders of S3

The only domains I’m keeping are burningbird.net, shelleypowers.com, and missourigreen.com. Burningbird is my major site, I’m turning the shelleypowers.com into an online CV, resume, what have you, and developing MissouriGreen more fully.

One unique feature of Missouri Green is that most of the site resources will be hosted on Amazon’s S3. I’ve already tried out S3 for a Flash-based photo show, and it works remarkably well. I figure that I can offload everything but the dynamic applications, such as whatever tool I use for the content. I’m leaning towards Drupal right now for the content. Drupal or WordPress most likely.

I’m using a variation of a Python script to bulk load to S3 from my server, but it needs work. I’m also using the Firefox add-on, S3 Firefox Organizer for loads from my desktop.

I’ve been enjoying myself immensely but I have to watch it: I’m almost up to a dollar in monthly charges.

Since S3 is a third party service, I’m not making it my key storage device. My photos I upload I have in RAW format on external hard drive and backup DVD. Same with the database dumps, as well as the code. If Amazon decides to enforce a minimum charge, or the service become less than robust, I have a plan to programatically recover the data and host elsewhere.

Categories
Diversity

The cultural divide

Kimberly Blessing has a good follow up discussion on the recent diversity discussion. She specifically pointed out something I also noticed, and it had to do with Robert Scoble’s comment to my post.

Robert wrote:

One thing about Digg and TechMeme (and, really, Megite and TailRank too): they reward networkers. How do you get links? Learn to beg for them via email and/or face-to-face meetings at conferences and other events. Men do this far far more often than women do.

Kimberly took issue with this because the expectation here is that we women have to emulate male behavior patterns in order to succeed. The whole point of diversity is that society, generally, and the tech field, specifically, has to work towards an environment that’s comfortable for all people, not just the current dominant holders of the gate. Otherwise, we lose all the benefits of diversity.

Contrary to what seems to be popular opinion among too many people, diversity is a good thing, not an onerous burden.

I agreed with Kimberly, but I also had an almost atavistic reaction to the whole idea of ‘begging’ for anything. I can’t speak for anyone else, but I know that I would never beg for anything. The very idea is hugely repugnant, with its reminders of not-so-long ago times when the only ‘rights’ given women were those which we could beg.

This isn’t to pick on Robert, but does demonstrate, perhaps, one of the many subtle and not so subtle ‘biases’ built into the tech environment. It is these cultural blips that makes the tech field fit like a tailored suit for the men, while women feel decidedly off the rack.

Categories
Writing

Short takes

The editing process can be fraught with possibilities for humor, especially when looking at typos. In one example, the text of the sample web site talked about “Poor the margarita into a glass”, leading to expressions of sympathy from my editors. My favorite so far, though, is “Transversing collections of data..” instead of “Traversing collections of data…”.

Alternate lifestyle code, indeed.

I’d already found the occurrence of “can’t” misspelled with a ‘u’ before sending it off for editing. Good thing, too, as editors never let you forget slips like this. Ever.

Speaking of geekery, OpenID is now the new cat’s jammies. For those new to this darling, all you really need is a bit of open source software and a URI, and it magically opens doors for you everywhere. I had an OpenID URI once but found it too blasè. I know what I really want, but unfortunately, iamgod.com, is already taken.

Categories
Critters

iCephalopod V

More cephaloporn thanks to Mr. Moult, who is also the happy owner of the einsteinslock.com domain.

The site linked has some wonderful, though short, videos of Taningia danae, eight armed squid, captured recently by Tsunemi Kubodera. They demonstrate how amazingly quick and aggressive squid can be. When you consider how big the colossal was that was recently recovered, 39 feet long, almost 12 meters, makes you think twice about going for a dip, eh?

At least I didn’t invoke Gorton’s Law and equate colossal with tire sized calamari:

In the tradition of Internet adages, I’d like to add another. In any discussion of sea life, no matter how rare, strange or disgusting, some knucklehead will always ask how well it goes with lemon or butter. I am calling this Gorton’s Law.

In case you haven’t been around, like, forever, I wrote a four part story on giant squid, Nessie, legends, and cryptozoology back in 1997, which is when I started developing a fascination of all things in our Other World. In celebration of its ten year anniversary, a link to the first part, A Tale of Two Monsters: Legends.

I wonder if I’ll get my annual nasty email from Jan?